Abstract

This document specifies VoiceXML, the Voice Extensible Markup Language. VoiceXML is designed for creating audio dialogs that feature synthesized speech, digitized audio, recognition of spoken and DTMF key input, recording of spoken input, telephony, and mixed-initiative conversations. Its major goal is to bring the advantages of web-based development and content delivery to interactive voice response applications.

Status of this Document

This specification describes markup for representing audio dialogs, and forms part of the proposals for the W3C Speech Interface Framework. This document has been produced as part of the W3C Voice Browser Activity, following the procedures set out for the W3C Process. The authors of this document are members of the Voice Browser Working Group (W3C Members only). This document is for public review, and comments and discussion are welcomed on the public mailing list <www-voice@w3.org>. To subscribe, send an email to <www-voice-request@w3. org> with the word subscribe in the subject line (include the word unsubscribe if you want to unsubscribe). The archive for the list is accessible online.

The proposed XML-based media types used in this specification have been submitted to the IETF for registration. Please note that during the registration process, the proposed media types may be modified or removed.

The Memorandum of Understanding between the W3C and the Voice XML Forum has paved the way for the publication of this working draft, with the VoiceXML Forum committing to abandoning trademark applications involving the name "VoiceXML".

This document seeks Member and public comment on both the technical design and the patent licensing issues arising out of the disclosure and licensing statements that have been made. Our decision to publish this first public working draft has been made to secure early comments from the community, but does not imply that all questions of patent licensing have been resolved or clarified. They must be resolved or work on this document in W3C will stop. As things stand at the time of publication of this specification, implementations conforming to this specification may require royalty bearing licenses for essential IPR. Further information can be found in the patent disclosures page. The patent policy for W3C as a whole is under wide discussion. A set of commitments by all participants in the Voice Browser Activity to royalty free is a possibility for the future but has NOT been made at time of publication.

This is a W3C Working Draft for review by W3C Members and other interested parties. It is a draft document and may be updated, replaced or made obsolete by other documents at any time. It is inappropriate to use W3C Working Drafts as reference material or to cite them as other than "work in progress".

This is work in progress and does not imply endorsement by the W3C membership. A list of current W3C Recommendations and other technical documents, including Working Drafts and Notes, can be found at http://www.w3.org/TR/.

Conventions of this Document

In this document, the key words "must", "must not", "required", "shall", "shall not", "should", "should not", "recommended", "may", and "optional" are to be interpreted as described in RFC 2119 and indicate requirement levels for compliant VoiceXML implementations.

Table of Contents

Abbreviated Contents

• 1. Overview
• 2. Dialog Constructs
• 3. User Input
• 4. System Output
• 5. Control flow and scripting
• 6. Environment and Resources
• 7. Appendices

Full Contents

• 1. Overview
  • 1.1 Introduction
  • 1.2 Background
    • 1.2.1 Architectural Model
    • 1.2.2 Goals of VoiceXML
    • 1.2.3 Scope of VoiceXML
    • 1.2.4 Principles of Design
    • 1.2.5 Implementation Platform Requirements
  • 1.3 Concepts
    • 1.3.1 Dialogs and Subdialogs
    • 1.3.2 Sessions
    • 1.3.3 Applications
    • 1.3.4 Grammars
    • 1.3.5 Events
    • 1.3.6 Links
  • 1.4 VoiceXML Elements
  • 1.5 Document Structure and Execution
    • 1.5.1 Execution within one Document
    • 1.5.2 Executing a Multi-Document Application
    • 1.5.3 Subdialogs
• 2. Dialog Constructs
  • 2.1 Forms
    • 2.1.1 Form Interpretation
    • 2.1.2 Form Items
    • 2.1.3 Form Item Variables and Conditions
    • 2.1.4 Directed Forms
    • 2.1.5 Mixed Initiative Forms
    • 2.1.6 Form Interpretation Algorithm
  • 2.2 Menus
  • 2.3 Form Items
    • 2.3.1 FIELD
    • 2.3.2 BLOCK
    • 2.3.3 INITIAL
    • 2.3.4 SUBDIALOG
    • 2.3.5 OBJECT
    • 2.3.6 RECORD
    • 2.3.7 TRANSFER
  • 2.4 Filled
  • 2.5 Links
• 3. User Input
  • 3.1 Grammars
    • 3.1.1 Speech Grammars
    • 3.1.2 DTMF Grammars
    • 3.1.3 Scope of Grammars
    • 3.1.4 Activation of Grammars
    • 3.1.5 Semantic Interpretation of Input
    • 3.1.6 Mapping Semantic Interpretation Results to VoiceXML forms
• 4. System Output
  • 4.1 Prompt
    • 4.1.1 Basic Prompts
    • 4.1.2 Speech Markup
    • 4.1.3 Audio Prompting
    • 4.1.4 <value> Element
    • 4.1.5 Barge-in
    • 4.1.6 Prompt Selection
    • 4.1.7 Timeout
    • 4.1.8 Prompt Queueing and Input Collection
• 5. Control flow and scripting
  • 5.1 Variables and Expressions
    • 5.1.1 Declaring Variables
    • 5.1.2 Variable Scopes
    • 5.1.3 Referencing Variables
    • 5.1.4 Standard Session Variables
    • 5.1.5 Standard Application Variables
  • 5.2 Event Handling
    • 5.2.1 Throw
    • 5.2.2 Catch
    • 5.2.3 Shorthand Notation
    • 5.2.4 Catch Element Selection
    • 5.2.5 Default Catch Elements
    • 5.2.6 Event Types
  • 5.3 Executable Content
    • 5.3.1 VAR
    • 5.3.2 ASSIGN
    • 5.3.3 CLEAR
    • 5.3.4 IF, ELSEIF, ELSE
    • 5.3.5 PROMPT
    • 5.3.6 REPROMPT
    • 5.3.7 GOTO
    • 5.3.8 SUBMIT
    • 5.3.9 EXIT
    • 5.3.10 RETURN
    • 5.3.11 DISCONNECT
    • 5.3.12 SCRIPT
    • 5.3.13 LOG
• 6. Environment and Resources
  • 6.1 Resource Fetching
    • 6.1.1 Fetching
    • 6.1.2 Caching
    • 6.1.3 Prefetching
    • 6.1.4 Protocols
  • 6.2 Meta
  • 6.3 Property
    • 6.3.1 Platform-Specific Properties
    • 6.3.2 Generic Speech Recognizer Properties
    • 6.3.3 Generic DTMF Recognizer Properties
    • 6.3.4 Prompt and Collect Properties
    • 6.3.5 Fetching Properties
    • 6.3.6 Miscellaneous Properties
  • 6.4 Param
  • 6.5 Time Designations
• 7. Appendices
  • Appendix A. Glossary of Terms
  • Appendix B. VoiceXML Document Type Definition
  • Appendix C. Form interpretation Algorithm
  • Appendix D. Timing Properties
  • Appendix E. Audio File Formats
  • Appendix F. Conformance
  • Appendix G. Internationalization
  • Appendix H. Accessibility
  • Appendix I. Privacy
  • Appendix J. Changes from VoiceXML 1.0
  • Appendix K. Reusability
  • Appendix L. Acknowledgements

1. Overview

This document defines VoiceXML, the Voice Extensible Markup Language. Its background, basic concepts and use are presented in Section 1. The dialog constructs of form, menu and link, and the mechanism (Form Interpretation Algorithm) by which they are interpreted are then introduced in Section 2. User input using DTMF and speech grammars is covered in Section 3, while Section 4 covers system output using speech synthesis and recorded audio. Mechanisms for manipulating dialog control flow, including variables, events, and executable elements, are explained in Section 5. Environment features such as parameters and properties as well as resource handling are specified in Section 6. The appendices provide additional information including the VoiceXML DTD, a detailed specification of the Form Interpretation Algorithm and timing, audio file formats, and statements relating to conformance, internationalization, accessibility and privacy.

Developers familar with VoiceXML 1.0 are particularly directed to Changes from Previous Public Version which summarizes how VoiceXML 2.0 differs from VoiceXML 1.0.

1.1 Introduction

VoiceXML is designed for creating audio dialogs that feature synthesized speech, digitized audio, recognition of spoken and DTMF key input, recording of spoken input, telephony, and mixed-initiative conversations. Its major goal is to bring the advantages of web-based development and content delivery to interactive voice response applications.

Here are two short examples of VoiceXML. The first is the venerable "Hello World":

<?xml version="1.0"?> 
<vxml version="2.0">
  <form>
    <block>Hello World!</block>
  </form>
</vxml>

The top-level element is <vxml>, which is mainly a container for dialogs. There are two types of dialogs: forms and menus. Forms present information and gather input; menus offer choices of what to do next. This example has a single form, which contains a block that synthesizes and presents "Hello World!" to the user. Since the form does not specify a successor dialog, the conversation ends.

Our second example asks the user for a choice of drink and then submits it to a server script:

<?xml version="1.0"?>
<vxml version="2.0">
  <form>
  <field name="drink">
     <prompt>Would you like coffee,tea, milk, or nothing?</prompt>
     <grammar src="drink.grxml" type="application/grammar+xml"/>
  </field>
  <block>
     <submit next="http://www.drink.example.com/drink2.asp"/>
  </block>
 </form>
</vxml>

A field is an input field. The user must provide a value for the field before proceeding to the next element in the form. A sample interaction is:

C (computer): Would you like coffee, tea, milk, or nothing?

H (human): Orange juice.

C: I did not understand what you said. (a platform-specific default message.)

C: Would you like coffee, tea, milk, or nothing?

H: Tea

C: (continues in document drink2.asp)

1.2 Background

This section contains a high-level architectural model, whose terminology is then used to describe the goals of VoiceXML, its scope, its design principles, and the requirements it places on the systems that support it.

1.2.1 Architectural Model

The architectural model assumed by this document has the following components:

Figure 1: Architectural Model

A document server (e.g. a web server) processes requests from a client application, the VoiceXML Interpreter, through the VoiceXML interpreter context. The server produces VoiceXML documents in reply, which are processed by the VoiceXML Interpreter. The VoiceXML interpreter context may monitor user inputs in parallel with the VoiceXML interpreter. For example, one VoiceXML interpreter context may always listen for a special escape phrase that takes the user to a high-level personal assistant, and another may listen for escape phrases that alter user preferences like volume or text-to-speech characteristics.

The implementation platform is controlled by the VoiceXML interpreter context and by the VoiceXML interpreter. For instance, in an interactive voice response application, the VoiceXML interpreter context may be responsible for detecting an incoming call, acquiring the initial VoiceXML document, and answering the call, while the VoiceXML interpreter conducts the dialog after answer. The implementation platform generates events in response to user actions (e.g. spoken or character input received, disconnect) and system events (e.g. timer expiration). Some of these events are acted upon by the VoiceXML interpreter itself, as specified by the VoiceXML document, while others are acted upon by the VoiceXML interpreter context.

1.2.2 Goals of VoiceXML

VoiceXML’s main goal is to bring the full power of web development and content delivery to voice response applications, and to free the authors of such applications from low-level programming and resource management. It enables integration of voice services with data services using the familiar client-server paradigm. A voice service is viewed as a sequence of interaction dialogs between a user and an implementation platform. The dialogs are provided by document servers, which may be external to the implementation platform. Document servers maintain overall service logic, perform database and legacy system operations, and produce dialogs. A VoiceXML document specifies each interaction dialog to be conducted by a VoiceXML interpreter. User input affects dialog interpretation and is collected into requests submitted to a document server. The document server replies with another VoiceXML document to continue the user’s session with other dialogs.

VoiceXML is a markup language that:

• Minimizes client/server interactions by specifying multiple interactions per document.

• Shields application authors from low-level, and platform-specific details.

• Separates user interaction code (in VoiceXML) from service logic (CGI scripts).

• Promotes service portability across implementation platforms. VoiceXML is a common language for content providers, tool providers, and platform providers.

• Is easy to use for simple interactions, and yet provides language features to support complex dialogs.

While VoiceXML strives to accommodate the requirements of a majority of voice response services, services with stringent requirements may best be served by dedicated applications that employ a finer level of control.

1.2.3 Scope of VoiceXML

The language describes the human-machine interaction provided by voice response systems, which includes:

• Output of synthesized speech (text-to-speech).

• Output of audio files.

• Recognition of spoken input.

• Recognition of DTMF input.

• Recording of spoken input.

• Control of dialog flow.

• Telephony features such as call transfer and disconnect.

The language provides means for collecting character and/or spoken input, assigning the input to document-defined request variables, and making decisions that affect the interpretation of documents written in the language. A document may be linked to other documents through Universal Resource Identifiers (URIs).

1.2.4 Principles of Design

VoiceXML is an XML application. For details about XML, refer to the Annotated XML Specification .

1. The language promotes portability of services through abstraction of platform resources.

2. The language accommodates platform diversity in supported audio file formats, speech grammar formats, and URI schemes. While producers of platforms may support various grammar formats the language requires a common grammar format, namely the XML Form of the W3C Speech Recognition Grammar Format, to facilitate interoperability. Similarly, while various audio formats for playback and recording may be supported, the audio formats described in Appendix E must be supported

3. The language supports ease of authoring for common types of interactions.

4. The language has a well-defined semantics that preserves the author's intent regarding the behavior of interactions with the user. Client heuristics are not required to determine document element interpretation.

5. The language has a control flow mechanism.

6. The language enables a separation of service logic from interaction behavior.

7. It is not intended for intensive computation, database operations, or legacy system operations. These are assumed to be handled by resources outside the document interpreter, e.g. a document server.

8. General service logic, state management, dialog generation, and dialog sequencing are assumed to reside outside the document interpreter.

9. The language provides ways to link documents using URIs, and also to submit data to server scripts using URIs.

10. VoiceXML provides ways to identify exactly which data to submit to the server, and which HTTP method (get or post) to use in the submittal.

11. The language does not require document authors to explicitly allocate and deallocate dialog resources, or deal with concurrency. Resource allocation and concurrent threads of control are to be handled by the implementation platform.

1.2.5 Implementation Platform Requirements

This section outlines the requirements on the hardware/software platforms that will support a VoiceXML interpreter.

Document acquisition. The interpreter context is expected to acquire documents for the VoiceXML interpreter to act on. The "http" URI protocol must be supported. In some cases, the document request is generated by the interpretation of a VoiceXML document, while other requests are generated by the interpreter context in response to events outside the scope of the language, for example an incoming phone call. When issuing document requests via http, the interpreter context identifies itself using the "User-Agent" header variable with the value "<name>/<version>", for example, "acme-browser/1.2"

Audio output. An implementation platform must support audio output using audio files and text-to-speech (TTS). The platform must be able to freely sequence TTS and audio output. Audio files are referred to by a URI. The language specifies a required set of audio file formats which must be supported (see Appendix E); additional audio file formats may also be supported.

Audio input. An implementation platform is required to detect and report character and/or spoken input simultaneously and to control input detection interval duration with a timer whose length is specified by a VoiceXML document.

• It must report characters (for example, DTMF) entered by a user. Platforms must support the DTMF grammar format described in XML Form of the W3C Speech Recognition Grammar Format. They should also support the DTMF grammar format described in Augmented BNF Form of the W3C Speech Recognition Grammar Format.

• It must be able to receive speech recognition grammar data dynamically. It must be able to use speech grammar data in the XML Form of the W3C Speech Recognition Grammar Format. It should be able to receive speech recognition grammar data in the Augmented BNF Form of the W3C Speech Recognition Grammar Format, and may support other formats such as the JSpeech Grammar Format or proprietary formats. Some VoiceXML elements contain speech grammar data; others refer to speech grammar data through a URI. The speech recognizer must be able to accommodate dynamic update of the spoken input for which it is listening through either method of speech grammar data specification.

• It must be able to record audio received from the user. The implementation platform must be able to make the recording available to a request variable. The language specifies a required set of recorded audio file formats which must be supported (see Appendix E); additional formats may also be supported.

Transfer The platform should be able to support making a third party connection through a communications network, such as the telephone.

1.3 Concepts

A VoiceXML document (or a set of documents called an application) forms a conversational finite state machine. The user is always in one conversational state, or dialog, at a time. Each dialog determines the next dialog to transition to. Transitions are specified using URIs, which define the next document and dialog to use. If a URI does not refer to a document, the current document is assumed. If it does not refer to a dialog, the first dialog in the document is assumed. Execution is terminated when a dialog does not specify a successor, or if it has an element that explicitly exits the conversation.

1.3.1 Dialogs and Subdialogs

There are two kinds of dialogs: forms and menus. Forms define an interaction that collects values for a set of field item variables. Each field may specify a grammar that defines the allowable inputs for that field. If a form-level grammar is present, it can be used to fill several fields from one utterance. A menu presents the user with a choice of options and then transitions to another dialog based on that choice.

A subdialog is like a function call, in that it provides a mechanism for invoking a new interaction, and returning to the original form. Variable instances, grammars, and state information are saved and are available upon returning to the calling document. Subdialogs can be used, for example, to create a confirmation sequence that may require a database query; to create a set of components that may be shared among documents in a single application; or to create a reusable library of dialogs shared among many applications.

1.3.2 Sessions

A session begins when the user starts to interact with a VoiceXML interpreter context, continues as documents are loaded and processed, and ends when requested by the user, a document, or the interpreter context.

1.3.3 Applications

An application is a set of documents sharing the same application root document. Whenever the user interacts with a document in an application, its application root document is also loaded. The application root document remains loaded while the user is transitioning between other documents in the same application, and it is unloaded when the user transitions to a document that is not in the application. While it is loaded, the application root document’s variables are available to the other documents as application variables, and its grammars can also be set to remain active for the duration of the application.

Figure 2 shows the transition of documents (D) in an application that share a common application root document (root).

Figure 2: Transitioning between documents in an application.

1.3.4 Grammars

Each dialog has one or more speech and/or DTMF grammars associated with it. In machine directed applications, each dialog’s grammars are active only when the user is in that dialog. In mixed initiative applications, where the user and the machine alternate in determining what to do next, some of the dialogs are flagged to make their grammars active (i.e., listened for) even when the user is in another dialog in the same document, or on another loaded document in the same application. In this situation, if the user says something matching another dialog’s active grammars, execution transitions to that other dialog, with the user’s utterance treated as if it were said in that dialog. Mixed initiative adds flexibility and power to voice applications.

1.3.5 Events

VoiceXML provides a form-filling mechanism for handling "normal" user input. In addition, VoiceXML defines a mechanism for handling events not covered by the form mechanism.

Events are thrown by the platform under a variety of circumstances, such as when the user does not respond, doesn't respond intelligibly, requests help, etc. The interpreter also throws events if it finds a semantic error in a VoiceXML document. Events are caught by catch elements or their syntactic shorthand. Each element in which an event can occur may specify catch elements. Catch elements are also inherited from enclosing elements "as if by copy". In this way, common event handling behavior can be specified at any level, and it applies to all lower levels.

1.3.6 Links

A link supports mixed initiative. It specifies a grammar that is active whenever the user is in the scope of the link. If user input matches the link’s grammar, control transfers to the link’s destination URI. A link can be used to throw an event to go to a destination URI.

1.4 VoiceXML Elements

1.5 Document Structure and Execution

A VoiceXML document is primarily composed of top-level elements called dialogs. There are two types of dialogs: forms and menus. A document may also have <meta> elements, <var> and <script> elements, <property> elements, <catch> elements, and <link> elements.

1.5.1 Execution within One Document

Document execution begins at the first dialog by default. As each dialog executes, it determines the next dialog. When a dialog doesn’t specify a successor dialog, document execution stops.

Here is "Hello World!" expanded to illustrate some of this. It now has a document level variable called "hi" which holds the greeting. Its value is used as the prompt in the first form. Once the first form plays the greeting, it goes to the form named "say_goodbye", which prompts the user with "Goodbye!" Because the second form does not transition to another dialog, it causes the document to be exited.

<?xml version="1.0"?>
<vxml version="2.0">
 <meta name="author" content="John Doe"/>
 <meta name="maintainer" content="hello-support@hi.example.com"/>
 <var name="hi" expr="'Hello World!'"/>
 <form>
  <block>
     <value expr="hi"/>
     <goto next="#say_goodbye"/>
  </block>
 </form>
 <form id="say_goodbye">
  <block>
     Goodbye!
  </block>
 </form>
</vxml>

Stylistically it is best to combine the forms:

<?xml version="1.0"?>
<vxml version="2.0">
 <meta name="author" content="John Doe"/>
 <meta name="maintainer" content="hello-support@hi.example.com"/>
 <var name="hi" expr="'Hello World!'"/>
 <form>
  <block>
     <value expr="hi"/> Goodbye!
  </block>
 </form>
</vxml>

Attributes of <vxml> include:

Language information is inherited down the document hierarchy: the value of "xml:lang" is inherited by elements which also define the "xml:lang" attribute, such as <grammar> and <prompt>, unless these elements specify an alternative value.

1.5.2 Executing a Multi-Document Application

Normally, each document runs as an isolated application. In cases where you want multiple documents to work together as one application, you select one document to be the application root document, and refer to it in the other documents’<vxml> elements.

When this is done, every time the interpreter is told to load a document in this application, it also loads the application root document if it is not already loaded. The application root document remains loaded until the interpreter is told to load a document that belongs to a different application. Thus one of the following two conditions always holds during interpretation:

• The application root document is loaded and the user is executing in it.

• The application root document and one other document in the application, known as an application leaf document, are both loaded and the user is executing in the leaf document.

If there is a chain of subdialogs defined in separate documents, then there may be more than one leaf document loaded although execution will only be in one of these documents.

There are several benefits to multi-document applications. First, the application root document’s variables are available for use by the other documents in the application, so that information can be shared and retained. Second, the grammars of the application root document are active even when the user is in other application documents, so that the user can always interact with forms, links, and menus from the application root document. Third, catch elements of the application root document can define default event handling for all documents within the application. Fourth, property elements in the application root document can specify default values for properties used throughout the application.

Here is a two-document application illustrating this:

Application root document (app-root.vxml)

<?xml version="1.0"?>
<vxml version="2.0">
 <var name="bye" expr="'Ciao'"/>
 <link next="operator_xfer.vxml">
   <grammar>
   <rule id="root" scope="public">operator</rule>
  </grammar>
 </link>
</vxml>

Leaf document (leaf.vxml)

<?xml version="1.0"?>
<vxml version="2.0" application="app-root.vxml">
 <form id="say_goodbye">
  <field name="answer" type="boolean">
     <prompt>Shall we say <value expr="application.bye"/>?</prompt>
     <filled>
       <if cond="answer">
        <exit/>
       </if>
       <clear namelist="answer"/>
     </filled>
  </field>
 </form>
</vxml>

In this example, the application is designed so that leaf.vxml must be loaded first. Its application attribute specifies that app-root.vxml should be used as the application root document. So, app-root.vxml is then loaded, which creates the application variable bye and also defines a link that navigates to operator-xfer.vxml whenever the user says "operator". The user starts out in the say_goodbye form:

C: Shall we say Ciao?

H: Si.

C: I did not understand what you said. (a platform-specific default message.)

C: Shall we say Ciao?

H: Ciao

C: I did not understand what you said.

H: Operator.

C: (Goes to operator_xfer.vxml, which transfers the caller to a human operator.)

Note that when the user is in a multi-document application, at most two documents are loaded at any one time: the application root document and, unless the user is actually interacting with the application root document, an application leaf document.A root document's <vxml> element does not have an application attribute specified. A leaf document's <vxml> element does have an application attribute specified. An interpreter always has an application root document loaded; it does not always have an application leaf document loaded.

The name of the interpreter's current application is the application root document's absolute URI. The absolute URI includes a query string, if present, but it does not include a fragment identifier. The interpreter remains in the same application as long as the name remains the same. When the name changes, a new application is entered and its root context is initialized. The application's root context consists of the variables, grammars, catch elements, and properties in application scope.

During a user session an interpreter transitions from one document to another as requested by <choice>, <goto> <link>, <subdialog>, and <submit> elements. Some transitions are within an application, others are between applications. The preservation or initialization of the root context depends on the type of transition:

Root to Leaf Within Application

A root to leaf transition within the same application occurs when the current document is a root document and the target document's application attribute's value resolves to the same absolute URI as the name of the current application. The application root document and its context are preserved.

Leaf to Leaf Within Application

A leaf to leaf transition within the same application occurs when the current document is a leaf document and the target document's application attribute's value resolves to the same absolute URI as the name of the current application. The application root document and its context are preserved.

Leaf to Root Within Application

A leaf to root transition within the same application occurs when the current document is a leaf document and the target document's absolute URI is the same as the name of the current application. The current application root document and its context are preserved when the transition is caused by a <choice>, <goto>, or <link> element. The root context is initialized when a <submit> element causes the leaf to root transition.

Root to Root

A root to root transition occurs when the current document is a root document and the target document is a root document, i.e. it does not have an application attribute. The root context is initialized with the new application root document, even if the documents are the same or have the same application name.

Subdialog

A subdialog invocation occurs when a root or leaf document executes a <subdialog> element. As discussed in Section 2.3.4, subdialog invocation creates a new execution context. The application root document and its context in the calling document's execution context are preserved untouched during subdialog execution, and are used again once the subdialog returns. A subdialog's new execution context has its own root context and, possibly, leaf context. If a subdialog is invoked with an empty URI reference and a fragment identifier, e.g. "#sub1", the new root and leaf contexts are initialized from the root and leaf documents used by the calling dialog. When the subdialog is invoked with a non-empty URI reference, the caching policy in Section 6.1.2 is used to acquire the root and leaf documents that will be used to initialize the new root and leaf contexts.

Inter-Application Transitions

All other transitions are between applications which cause the application root context to be initialized with the next application's root document.

If a document refers to a non-existent application root document, or if a document's application attribute refers to a document that also has an application attribute specified, an error.semantic event is thrown.

The following diagrams illustrate the effect of the transitions between root and leaf documents on the application root context. In these diagrams, boxes represent documents, box texture changes identify root context initialization, solid arrows symbolize transitions to the URI in the arrow's label, dashed vertical arrows indicate an application attribute whose URI is the arrow's label.

Figure 3: Transitions that Preserve the Root Context

In this diagram, all the documents belong to the same application. The transitions are identified by the numbers 1-4 across the top of the figure. They are:

1. A transition to URI A results in document 1, the application context is initialized from document 1's content. Assume that this is the first document in the session. The current application's name is A.
2. Document 1 specifies a transition to URI B, which yields document 2. Document 2's application attribute equals URI A. The root is document 1 with its context preserved. This is a root to leaf transition within the same application.
3. Document 2 specifies a transition to URI C, which yields another leaf document, document 3. Its application attribute also equals URI A. The root is document 1 with its context preserved. This is a leaf to leaf transition within the same application.
4. Document 3 specifies a transition to URI A using a <choice>, <goto>, or <link>. Document 1 is used with its root context intact. This is a leaf to root transition within the same application.

The next diagram illustrates transitions which initialize the root context.

Figure 4: Transitions that Initialize the Root Context

5. Document 1 specifies a transition to its own URI A. The resulting document 4 does not have an application attribute, so it is considered a root document, and the root context is initialized. This is a root to root transition.
6. Document 4 specifies a transition to URI D, which yields a leaf document 5. Its application attribute is different: URI E. A new application is being entered. URI E produces the root document 6. The root context is initialized from the content of document 6. This is an inter-application transition.
7. Document 5 specifies a transition to URI A. The cache check returns document 4 which does not have an application attribute and therefore belongs to application A, so the root context is initialized. Initialization occurs even though this application and this root document were used earlier in the session. This is an inter-application transition.

1.5.3 Subdialogs

A subdialog is a mechanism for decomposing complex sequences of dialogs to better structure them, or to create reusable components. For example, the solicitation of account information may involve gathering several pieces of information, such as account number, and home telephone number. A customer care service might be structured with several independent applications that could share this basic building block, thus it would be reasonable to construct it as a subdialog. This is illustrated in the example below. The first document, app.vxml, seeks to adjust a customer’s account, and in doing so must get the account information and then the adjustment level. The account information is obtained by using a subdialog element that invokes another VoiceXML document to solicit the user input. While the second document is being executed, the calling dialog is suspended, awaiting the return of information. The second document provides the results of its user interactions using a <return> element, and the resulting values are accessed through the variable defined by the name attribute on the <subdialog> element.

Customer Service Application (app.vxml)

<?xml version="1.0"?>
<vxml version="2.0">
  <form id="billing_adjustment">
    <var name="account_number"/>
    <var name="home_phone"/>
    <subdialog name="accountinfo" src="acct_info.vxml#basic">
      <filled>
        <!-- Note the variable defined by "accountinfo" is
          returned as an ECMAScript object and it contains two
          properties defined by the variables specified in the
          "return" element of the subdialog. -->

        <assign name="account_number" expr="accountinfo.acctnum"/>
        <assign name="home_phone" expr="accountinfo.acctphone"/>
      </filled>
    </subdialog>

    <field name="adjustment_amount" type="currency">
      <prompt>
        What is the value of your account adjustment?
      </prompt>
      <filled>
        <submit  next="/cgi-bin/updateaccount"/>
      </filled>
    </field>
  </form>
</vxml>

Document Containing Account Information Subdialog (acct_info.vxml)

<?xml version="1.0"?>
<vxml version="2.0">
  <form id="basic">
    <field name="acctnum" type="digits">
      <prompt> What is your account number? </prompt>
    </field>
    <field name="acctphone" type="phone">
      <prompt> What is your home telephone number? </prompt>

      <filled>
        <!-- The values obtained by the two fields are supplied
          to the calling dialog by the "return" element. -->
        <return namelist="acctnum acctphone"/>
      </filled>
    </field>
  </form>
</vxml>

Subdialogs add a new execution context when they are invoked.The subdialog could be a new dialog within the existing document, or a new dialog within a new document.

Subdialogs can be composed of several documents. Figure 5 shows the execution flow where a sequence of documents (D) transitions to a subdialog (SD) and then back.

Figure 5: Subdialog composed of several documents
returning from the last subdialog document.

The execution context in dialog D2 is suspended when it invokes the subdialog SD1 in document sd1.vxml. This subdialog specifies execution is to be transfered to the dialog in sd2.vxml (using <goto>). Consequently, when the dialog in sd2.vxml returns, control is returned directly to dialog D2.

Figure 6 shows an example of a multi-document subdialog where control is transferred from one subdialog to another.

Figure 6: Subdialog composed of several documents
returning from the first subdialog document.

The subdialog in sd1.vxml specifies that control is to be transfered to a second subdialog, SD2, in sd2.vxml. When executing SD2, there are two suspended contexts: the dialog context in D2 is suspending awaiting SD1 to return; and the dialog context in SD1 awaiting SD2 to return. When SD2 returns, control is returned to the SD1. It in turn returns control to dialog D2.

2. Dialog Constructs

2.1 Forms

Forms are the key component of VoiceXML documents. A form contains:

• A set of form items, elements that are visited in the main loop of the form interpretation algorithm. Form items are subdivided into field items, those that define the form’s field item variables, and control items, those that help control the gathering of the form’s fields.

• Declarations of non-field item variables.

• Event handlers.

• "Filled" actions, blocks of procedural logic that execute when certain combinations of field items are filled in.

Form attributes are:

This section describes some of the concepts behind forms, and then gives some detailed examples of their operation.

2.1.1 Form Interpretation

Forms are interpreted by an implicit form interpretation algorithm (FIA). The FIA has a main loop that repeatedly selects a form item and then visits it. The selected form item is the first in document order whose guard condition is not satisfied. For instance, a field item’s default guard condition tests to see if the field item variable has a value, so that if a simple form contains only field items, the user will be prompted for each field item in turn.

Interpreting a form item generally involves:

• Selecting and playing one or more prompts;

• Collecting a user input, either a response that fills in one or more fields, or a throwing of some event (help, for instance); and

• Interpreting any <filled> actions that pertained to the newly filled in fields.

The FIA ends when it interprets a transfer of control statement (e.g. a <goto> to another dialog or document, or a <submit> of data to the document server). It also ends with an implied <exit> when no form item remains eligible to select.

The FIA is described in more detail in Section 2.1.6.

2.1.2 Form Items

A form’s form items are the elements that can be visited in the main loop of the form interpretation algorithm. Field items direct the FIA to gather a specific field. When the FIA selects a control item, the control item may contain a block of procedural code to execute, or it may tell the FIA to set up the initial prompt-and-collect for a mixed initiative form.

2.1.2.1 Field Items

A field item specifies a field item variable to gather from the user. Field items have prompts to tell the user what to say or key in, grammars that define the allowed inputs, and event handlers that process any resulting events. A field item may also have a <filled> element that defines an action to take just after the field item variable is filled in. Field items are subdivided into:

2.1.2.2 Control Items

There are two types of control items:

2.1.3 Form Item Variables and Conditions

Each form item has an associated form item variable, which by default is set to undefined when the form is entered. This form item variable will contain the result of interpreting the form item. A field item’s form item variable is also called a field item variable, and it holds the value collected from the user. A form item variable can be given a name using the name attribute, or left nameless, in which case an internal name is generated.

Each form item also has a guard condition, which governs whether or not that form item can be selected by the form interpretation algorithm. The default guard condition just tests to see if the form item variable has a value. If it does, the form item will not be visited.

Typically, field items are given names, but control items are not. Generally form item variables are not given initial values and additional guard conditions are not specified. But sometimes there is a need for more detailed control. One form may have a form item variable initially set to hide a field, and later cleared (e.g., using <clear>) to force the field’s collection. Another field may have a guard condition that activates it only when it has not been collected, and when two other fields have been filled. A block item could execute only when some condition holds true. Thus, fine control can be exercised over the order in which form items are selected and executed by the FIA, however in general, many dialogs can be constructed without resorting to this level of complexity.

In summary, all form items have the following attributes:

2.1.4 Directed Forms

The simplest and most common type of form is one in which the form items are executed exactly once in sequential order to implement a computer-directed interaction. Here is a weather information service that uses such a form.

<form id="weather_info">
 <block>Welcome to the weather information service.</block>
 <field name="state">
  <prompt>What state?</prompt>
  <grammar src="state.grxml"  type="application/grammar+xml"/>
  <catch event="help">
     Please speak the state for which you want the weather.
  </catch>
 </field>
 <field name="city">
  <prompt>What city?</prompt>
  <grammar src="city.grxml" type="application/grammar+xml"/>
  <catch event="help">
     Please speak the city for which you want the weather.
  </catch>
 </field>
 <block>
  <submit next="/servlet/weather" namelist="city state"/>
 </block>
</form>

This dialog proceeds sequentially:

C (computer): Welcome to the weather information service. What state?

H (human): Help

C: Please speak the state for which you want the weather.

H: Georgia

C: What city?

H: Tblisi

C: I did not understand what you said. What city?

H: Macon

C: The conditions in Macon Georgia are sunny and clear at 11 AM ...

The form interpretation algorithm’s first iteration selects the first block, since its (hidden) form item variable is initially undefined. This block outputs the main prompt, and its form item variable is set to true. On the FIA’s second iteration, the first block is skipped because its form item variable is now defined, and the state field is selected because the dialog variable state is undefined. This field prompts the user for the state, and then sets the variable state to the answer. The third form iteration prompts and collects the city field. The fourth iteration executes the final block and transitions to a different URI.

Each field in this example has a prompt to play in order to elicit a response, a grammar that specifies what to listen for, and an event handler for the help event. The help event is thrown whenever the user asks for assistance. The help event handler catches these events and plays a more detailed prompt.

Here is a second directed form, one that prompts for credit card information:

<form id="get_card_info">
 <block>We now need your credit card type, number,
    and expiration date.</block>
 <field name="card_type">
  <prompt count="1">What kind of credit card
    do you have?</prompt>
  <prompt count="2">Type of card?</prompt>
  <!-- This is an inline grammar. -->
  <grammar>
    <rule id="r2" scope="public">
       <one-of>
       <item>visa</item>
       <item>master <count number="optional">card</count></item>
       <item>amex</item>
       <item>american express</item>
       </one-of>
    </rule>
  </grammar>
  <help> Please say Visa, Mastercard, or American Express.</help>
 </field>

 <!-- The grammar for type="digits" is built in. -->
 <field name="card_num" type="digits">
  <prompt count="1">What is your card number?</prompt>
  <prompt count="2">Card number?</prompt>
  <catch event="help">
  <if cond="card_type =='amex' || card_type =='american express'">
       Please say or key in your 15 digit card number.
     <else/>
       Please say or key in your 16 digit card number.
     </if>
  </catch>
  <filled>
<if cond="(card_type == 'amex' || card_type =='american express') 
          && card_num.length != 15">
       American Express card numbers must have 15 digits.
       <clear namelist="card_num"/>
       <throw event="nomatch"/>
     <elseif cond="card_type != 'amex'
                     && card_type !='american express'
                     && card_num.length != 16"/>
       Mastercard and Visa card numbers have 16 digits.
       <clear namelist="card_num"/>
       <throw event="nomatch"/>
     </if>
  </filled>
 </field>

 <field name="expiry_date" type="digits">
   <prompt count="1">What is your card's expiration date?</prompt>
   <prompt count="2">Expiration date?</prompt>
  <help>
     Say or key in the expiration date, for example one two oh one.
  </help>
  <filled>
     <!-- validate the mmyy -->
     <var name="mm"/>
     <var name="i" expr="expiry_date.length"/>
     <if cond="i == 3">
       <assign name="mm" expr="expiry_date.substring(0,1)"/>
     <elseif cond="i == 4"/>
       <assign name="mm" expr="expiry_date.substring(0,2)"/>
     </if>
     <if cond="mm == '' || mm < 1 || mm > 12">
       <clear namelist="expiry_date"/>
       <throw event="nomatch"/>
     </if>
  </filled>
 </field>

 <field name="confirm" type="boolean">

  <prompt>
      I have <value expr="card_type"/> number
      <value expr="card_num"/>, expiring on
      <value expr="expiry_date"/>.
      Is this correct?
  </prompt>
  <filled>
    <if cond="confirm">
      <submit next="place_order.asp"
        namelist="card_type card_num expiry_date"/>
    </if>
    <clear namelist="card_type card_num expiry_date acknowledge"/>
  </filled>
 </field>
</form>

Note that the grammar alteratives 'amex' and 'american express' return literal values which need to be handled separately in the conditional expressions. Section 3.1.5 describes how semantic attachments in the grammar can be used to return a single representation of these inputs.

The dialog might go something like this:

C: We now need your credit card type, number, and expiration date.

C: What kind of credit card do you have?

H: Discover

C: I did not understand what you said. (a platform-specific default message.)

C: Type of card? (the second prompt is used now.)

H: Shoot. (fortunately treated as "help" by this platform)

C: Please say Visa, Master card, or American Express.

H: Uh, Amex. (this platform ignores "uh")

C: What is your card number?

H: One two three four ... wait ...

C: I did not understand what you said.

C: Card number?

H: (uses DTMF) 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 #

C: What is your card’s expiration date?

H: one two oh one

C: I have Amex number 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 expiring on 1 2 0 1. Is this correct?

H: Yes

Fields are the major building blocks of forms. A field declares a variable and specifies the prompts, grammars, DTMF sequences, help messages, and other event handlers that are used to obtain it. Each field declares a VoiceXML field item variable in the form’s dialog scope. These may be submitted once the form is filled, or copied into other variables.

Each field has its own speech and/or DTMF grammars, specified explicitly using <grammar> elements, or implicitly using the type attribute. The type attribute is used for standard builtin grammars, like digits, boolean, or number. The type attribute also governs how that field’s value is spoken by the speech synthesizer.

Each field can have one or more prompts. If there is one, it is repeatedly used to prompt the user for the value until one is provided. If there are many, they must be given count attributes. These determine which prompt to use on each attempt. In the example, prompts become shorter. This is called tapered prompting.

The <catch event="help"> elements are event handlers that define what to do when the user asks for help. Help messages can also be tapered. These can be abbreviated, so that the following two elements are equivalent:

<catch event="help">
 Please say visa, mastercard, or amex.
</catch>

<help>
 Please say visa, mastercard, or amex.
</help>

The <filled> element defines what to do when the user provides a recognized input for that field. One use is to specify integrity constraints over and above the checking done by the grammars, as with the date field above.

2.1.5 Mixed Initiative Forms

The last section talked about forms implementing rigid, computer-directed conversations. To make a form mixed initiative, where both the computer and the human direct the conversation, it must have one or more <initial> form items and one or more form-level grammars.

If a form has form-level grammars:

• Its fields can be filled in any order.

• More than one field can be filled as a result of a single user utterance.

The filling of field variables when using a form-level grammar is described in Section 3.1.6.

Also, the form’s grammars can be active when the user is in other dialogs. If a document has two forms on it, say a car rental form and a hotel reservation form, and both forms have grammars that are active for that document, a user could respond to a request for hotel reservation information with information about the car rental, and thus direct the computer to talk about the car rental instead. The user can speak to any active grammar, and have fields set and actions taken in response.

Example. Here is a second version of the weather information service, showing mixed initiative. It has been "enhanced" for illustrative purposes with advertising and with a confirmation of the city and state:

<form id="weather_info">

<grammar src="cityandstate.grxml" type="application/grammar+xml"/>

 <!-- Caller can't barge in on today's advertisement. -->
 <block>
  <prompt bargein="false">
     Welcome to the weather information service.
     <audio src="http://www.online-ads.example.com/wis.wav"/>
  </prompt>
 </block>

 <initial name="start">
  <prompt>
     For what city and state would you like the weather?
  </prompt>
  <help> 
    Please say the name of the city and
    state for which you would like a weather report.
  </help>
  <!-- If user is silent, reprompt once, then
   try directed prompts. -->
  <noinput count="1"> <reprompt/></noinput>
  <noinput count="2"> <reprompt/>
    <assign name="start" expr="true"/></noinput>
 </initial>

 <field name="state">
  <prompt>What state?</prompt>
  <help>
    Please speak the state for which you want the weather.
  </help>
 </field>

 <field name="city">
  <prompt>Please say the city in <value expr="state"/> 
                for which you want the weather.</prompt>
  <help>Please speak the city for which you
    want the weather.</help>
  <filled>
     <!-- Most of our customers are in LA. -->
     <if cond="city == 'Los Angeles' && state == undefined">

       <assign name="state" expr="'California'"/>
     </if>
  </filled>
 </field>

 <field name="go_ahead" type="boolean" modal="true">
  <prompt>Do you want to hear the weather for
    <value expr="city"/>, <value expr="state"/>?
  </prompt>
  <filled>
     <if cond="go_ahead">
       <prompt bargein="false">
         <audio src="http://www.online-ads.example.com/wis2.wav"/>
       </prompt>
       <submit next="/servlet/weather" namelist="city state"/>
     </if>
     <clear namelist="start city state go_ahead"/>
  </filled>
 </field>
</form>

Here is a transcript showing the advantages for even a novice user:

C: Welcome to the weather information service. Buy Joe’s Spicy Shrimp Sauce.

C: For what city and state would you like the weather?

H: Uh, California.

C: Please say the city in California for which you want the weather.

H: San Francisco, please.

C: Do you want to hear the weather for San Francisco, California?

H: No

C: For what city and state would you like the weather?

H: Los Angeles.

C: Do you want to hear the weather for Los Angeles, California?

H: Yes

C: Don’t forget, buy Joe’s Spicy Shrimp Sauce tonight!

C: Mostly sunny today with highs in the 80s. Lows tonight from the low 60s ...

The go_ahead field has its modal attribute set to true. This causes all grammars to be disabled except the ones defined in the current form item, so that the only grammar active during this field is the builtin grammar for boolean.

An experienced user can get things done much faster (but is still forced to listen to the ads):

C: Welcome to the weather information service. Buy Joe’s Spicy Shrimp Sauce.

C: What ...

H (barging in): LA

C: Do you ...

H (barging in): Yes

C: Don’t forget, buy Joe’s Spicy Shrimp Sauce tonight!

C: Mostly sunny today with highs in the 80s. Lows tonight from the low 60s ...

2.1.5.1 Controlling the order of field collection.

The form interpretation algorithm can be customized in several ways. One way is to assign a value to a form item variable, so that its form item will not be selected. Another is to use <clear> to set a form item variable to undefined; this forces the FIA to revisit the form item again.

Another method is to explicitly specify the next field item to visit using <goto nextitem>. This forces an immediate transfer to that field item. No variables, conditions or counters in the targeted form item will be reset. The form item's prompt will be played even if it has already been visited. If the <goto nextitem> occurs in a <filled> action, the rest of the <filled> action and any pending <filled> actions will be skipped.

Here is an example <goto nextitem> executed in response to the exit event:

<form id="survey_2000_03_30">
  <catch event="exit">
    <reprompt/>
    <goto nextitem="confirm_exit"/>
  </catch>
  <block>
   <prompt>
     Hello, you have been called at random to answer questions
     critical to U.S. foreign policy.
   </prompt>
  </block>

  <field name="q1" type="boolean">
   <prompt>Do you agree with the IMF position on
     privatizing certain functions of Burkina Faso’s
     agriculture ministry?</prompt>
  </field>

  <field name="q2" type="boolean">
    <prompt>If this privatization occurs, will its
      effects be beneficial mainly to Ouagadougou and
      Bobo-Dioulasso?</prompt>
  </field>

  <field name="q3" type="boolean">
    <prompt>Do you agree that sorghum and millet output
      might thereby increase by as much as four percent per
      annum?</prompt>
  </field>

  <block>
    <submit next="register" namelist="q1 q2 q3"/>
  </block>

  <field name="confirm_exit" type="boolean">
    <prompt>You have elected to exit.  Are you
      sure you want to do this, and perhaps adversely affect
      U.S. foreign policy vis-`-vis sub-Saharan Africa for
      decades to come?</prompt>
    <filled>
      <if cond="confirm_exit">
        Okay, but the U.S. State Department is displeased.
        <exit/>
      <else/>
        Good, let's pick up where we left off.
        <clear namelist="confirm_exit"/>
      </if>
    </filled>
  </field>
</form>

If the user says "exit" in response to any of the survey questions, an exit event is thrown by the platform and caught by the <catch> event handler. This handler directs that confirm_exit be the next visited field. The confirm_exit field would not be visited during normal completion of the survey because the preceding <block> element transfers control to the registration script.

2.1.6 Form Interpretation Algorithm

We’ve presented the form interpretation algorithm (FIA) at a conceptual level. In this section we describe it in more detail. A more formal description is provided in Appendix C.

2.1.6.1 Initialization Phase

Whenever a form is entered, it is initialized. Internal prompt counter variables (in the form’s dialog scope) are reset to 1. Each variable (form-level <var> elements and form item variables) is initialized, in document order, to undefined or to the value of the relevant expr attribute.

2.1.6.2 Main Loop

The main loop of the FIA has three phases:

The select phase: the next form item is selected for visiting.

The collect phase: the next unfilled form item is visited, which prompts the user for input, enables the appropriate grammars, and then waits for and collects an input (such as a spoken phrase or DTMF key presses) or an event (such as a request for help or a no input timeout).

The process phase: an input is processed by filling form items and executing <filled> elements to perform actions such as input validation. An event is processed by executing the appropriate event handler for that event type.

Note that the FIA may be given an input (a set of grammar slot/slot value pairs) that was collected while the user was in a different form’s FIA. In this case the first iteration of the main loop skips the select and collect phases, and goes right to the process phase with that input.

2.1.6.2.1 Select phase

The purpose of the select phase is to select the next form item to visit. This is done as follows:

If a <goto> from the last main loop iteration’s process phase specified a <goto nextitem>, then the specified form item is selected.

Otherwise the first form item whose guard condition is false is chosen to be visited.

If no guard condition is false, then the last iteration completed the form without encountering an explicit transfer of control, so the FIA does an implicit <exit> operation.

2.1.6.2.2 Collect phase

The purpose of the collect phase is to collect an input or an event. The selected form item is visited, which performs actions that depend on the type of form item:

If a field item is visited, the FIA selects and queues up any prompts based on the field item’s prompt counter and the prompt conditions. Then it listens for the field level grammar(s) and any active higher-level grammars, and waits for a grammar recognition or for some event.

If a <transfer> is visited, the prompts are queued based on the item’s prompt counter and the prompt conditions. The item grammars are activated. The queue is played before the transfer is executed.

If a <subdialog> or <object> is visited, the prompts are queued based on the item’s prompt counter and the prompt conditions. Grammars are not activated. Instead, the input collection behavior is specified by the executing context for the subdialog or object. The queue is not played before the subdialog or object is executed, but instead should be played during the subsequent input collection.

If an <initial> is visited, the FIA selects and queues up prompts based on the <initial>’s prompt counter and prompt conditions. Then it listens for the form level grammar(s) and any active higher-level grammars. It waits for a grammar recognition or for an event.

A <block> element is visited by setting its form item variable to true, evaluating its content, and then bypassing the process phase. No input is collected, and the next iteration of the FIA’s main loop is entered.

2.1.6.2.3 Process phase

The purpose of the process phase is to process the input or event collected during the collect phase, as follows:

• If an event (such as a noinput or a hangup) occurred, then the applicable catch element is identified and executed. Selection of the applicable catch element starts in the scope of the current form item and then proceeds outward by enclosing dialog scopes. This can cause the FIA to terminate (e.g. if it transitions to a different dialog or document or it does an <exit>), or it can cause the FIA to go into the next iteration of the main loop (e.g. as when the default help event handler is executed).
• If an input matches a grammar from a <link> then that link’s transition is executed, or its event is thrown. If the <link> throws an event, the event is processed in the context of the current form item (e.g. <initial>, <field>, <transfer>, and so forth).
• If an input matches a grammar in a form other than the current form, then the FIA terminates, the other form is initialized, and that form’s FIA is started with this input in its process phase.

If an input matches a grammar in this form, then:

• The input’s grammar slot values are assigned to the corresponding field item variables.
• The <filled> actions triggered by these assignments are identified as described in Section 2.4.
• Each identified <filled> action is executed in document order. If a <submit>, <disconnect>, <exit>, <return>, <goto> or <throw> is encountered, the remaining <filled> elements are not executed, and the FIA either terminates or continues in the next main loop iteration. <reprompt> does not terminate the FIA (the name suggests an action), but rather just sets a flag that affects the treatment of prompts on the subsequent iteration of the FIA. If an event is thrown during the execution of a <filled>, event handler selection starts in the scope of the <filled>, which could be a form item or the form itself, and then proceeds outward by enclosing dialog scopes.

After completion of the process phase, interpretation continues by returning to the select phase.

A more detailed form interpretation algorithm can be found in Appendix C.

2.2 Menus

A menu is a convenient syntactic shorthand for a form containing a single anonymous field that prompts the user to make a choice and transitions to different places based on that choice. Like a regular form, it can have its grammar scoped such that it is active when the user is executing another dialog. The following menu offers the user three choices:

<menu>
  <prompt>
    Welcome home. Say one of: <enumerate/>
  </prompt>
  <choice next="http://www.sports.example.com/vxml/start.vxml">
     Sports
  </choice>
  <choice next="http://www.weather.example.com/intro.vxml">
     Weather
  </choice>
  <choice next="http://www.stargazer.example.com/voice/astronews.vxml">
     Stargazer astrophysics news
  </choice>
  <noinput>Please say one of <enumerate/></noinput>
</menu>

This dialog might proceed as follows:

C: Welcome home. Say one of: sports; weather; Stargazer astrophysics news.

H: Astrology.

C: I did not understand what you said. (a platform-specific default message.)

C: Welcome home. Say one of: sports; weather; Stargazer astrophysics news.

H: sports.

C: (proceeds to http://www.sports.example.com/vxml/start.vxml)

Menu element

This identifies the menu, and determines the scope of its grammars. Menu attributes are:

Choice element

The <choice> element serves several purposes:

• It specifies a speech grammar fragment and/or a DTMF grammar fragment that determines when that choice has been selected.

• The contents are used to form the <enumerate> prompt string.

• It specifies the URI to go to when the choice is selected.

Choice attributes are:

If a <grammar> element is specified in <choice>, then the external grammar is used instead of an automatically generated grammar. This allows the developer to precisely control the <choice> grammar; for example:

<menu>
  <choice next="http://www.sports.example.com/vxml/start.vxml">
  <grammar  src="sports.grxml" type="application/grammar+xml"/>
  Sports
  </choice>
  <choice next="http://www.weather.example.com/intro.vxml">
  <grammar  src="weather.grxml" type="application/grammar+xml"/>
  Weather
  </choice>
  <choice next="http://www.stargazer.example.com/voice/astronews.vxml">
  <grammar  src="astronews.grxml" type="application/grammar+xml"/>
  Stargazer astrphysics
  </choice>
</menu>

DTMF in Menus

Menus can rely purely on speech, purely on DTMF, or both in combination by including a <property> element in the <menu>. Here is a DTMF-only menu with explicit DTMF sequences given to each choice, using the choice’s dtmf attribute:

<menu>
  <property name="inputmodes" value="dtmf"/>
  <prompt>
   For sports press 1, For weather press 2, For Stargazer
   astrophysics press 3.
  </prompt>
  <choice dtmf="1" next="http://www.sports.example.com/vxml/start.vxml"/>
  <choice dtmf="2" next="http://www.weather.example.com/intro.vxml"/>
  <choice dtmf="3" next="http://www.stargazer.example.com/astronews.vxml"/>
</menu>

Alternatively, you can set the <menu>’s dtmf attribute to true to assign sequential DTMF digits to each of the first nine choices: the first choice has DTMF "1", and so on:

<menu dtmf="true">
  <property name="inputmodes" value="dtmf"/>
  <prompt>
   For sports press 1, For weather
   press 2, For Stargazer astrophysics press 3.
  </prompt>
  <choice next="http://www.sports.example.com/vxml/start.vxml"/>
  <choice next="http://www.weather.example.com/intro.vxml"/>
  <choice
    next="http://www.stargazer.example.com/voice/astronews.vxml"/>
</menu>

Enumerate element

The <enumerate> element is an automatically generated description of the choices available to the user. It specifies a template that is applied to each choice in the order they appear in the menu. If it is used with no content, a default template that lists all the choices is used, determined by the interpreter context. If it has content, the content is the template specifier. This specifier may refer to two special variables: _prompt is the choice’s prompt, and _dtmf is the choice’s assigned DTMF sequence. For example, if the menu were rewritten as

<menu dtmf="true">
  <prompt>
    Welcome home.
    <enumerate>
     For <value expr="_prompt"/>, press <value
     expr="_dtmf"/>.
   </enumerate>
   </prompt>
   <choice next="http://www.sports.example.com/vxml/start.vxml">
      sports </choice>
   <choice next="http://www.weather.example.com/intro.vxml">
      weather </choice>
   <choice next="http://www.stargazer.example.com/voice/astronews.vxml">
      Stargazer astrophysics news
   </choice>
</menu>

then the menu’s prompt would be:

C: Welcome home. For sports, press 1. For weather, press 2. For Stargazer astrophysics news, press 3.

The <enumerate> element may be used within the prompts and the catch elements associated with <menu> elements and with <field> elements that contain <option> elements, as discussed in Section 2.3.1.3. An error.semantic event is thrown if <enumerate> is used elsewhere.

Grammar Generation

Any choice phrase specifies a set of words and phrases to listen for. A choice phrase is constructed from the PCDATA of the elements contained directly or indirectly in the <choice> element.

If the accept attribute is "exact" then the user must say the entire phrase in the same order in which they occur in the choice phrase.

If the accept attribute is "approximate", then the choice may be matched when a user says a subphrase of the expression. For example, in response to the prompt "Stargazer astrophysics news" a user could say "Stargazer", "astrophysics", "Stargazer news", "astrophysics news", and so on. The equivalent grammar may be language and platform dependent.

As an example of using "exact" and "approximate" in different choices, consider this example:

<menu accept="approximate">
  <choice next="..."> Stargazer Astrophysics News </choice>
  <choice accept="exact" next="..."> Physics Weekly  </choice>
  <choice accept="exact" next="..."> Particle Physics Update </choice>
  <choice next="..."> Astronomy Today </choice>
</menu>

Because "approximate" is specified for the first choice, the user may say a subphrase when matching the first choice; for instance, "Stargazer" or "Astrophysics News". However, because "exact" is specified in the second and third choices, only a complete phrase will match: "Physics Weekly" and "Partical Physics Update".

As an example of the use of PCDATA contained in descendants of the <choice> element, consider the following example:

<choice accept="exact"
next="http://www.stargazer.example.com/voice/astronews.vxml">
   <audio src="http://www.stargazer.example.com/space.wav">
     Stargazer <emphasis>astrophysics</emphasis> news
   </audio>
</choice>

This choice would be read from the audio file, or as "Stargazer Astrophysics News" if the file could not be played. The grammar for the choice would be the exact phrase "Stargazer astrophysics news" gleaned from the PCDATA of the <choice> element’s descendants.

Interpretation model

A menu behaves like a form with a single field that does all the work. The menu prompts become field prompts. The menu event handlers become the field event handlers. The menu grammars become form grammars.

Upon entry, the menu’s grammars are built and enabled, and the prompt is played. When the user input matches a choice, control transitions according to the value of the next, expr, or event attribute of the <choice>, only one of which may be specified. If an event attribute is specified but its event handler does not cause the interpreter to exit or transition control, then the FIA will clear the form item variable of the menu's anonymous field, causing the menu to be executed again.

2.3 Form Items

A form item is an element of a <form> that can be visited during form interpretation. They include <field>, <block>, <initial>, <subdialog>, <object>, <record>, and <transfer>.

All form items have the following characteristics:

• They have a result variable, specified by the name attribute. This variable may be given an initial value with the expr attribute.

• They have a guard condition specified with the cond attribute.

• Form items are subdivided into field items, those that define the form’s field item variables, and control items, those that help control the gathering of the form’s fields. Field items (<field>, <subdialog>, <object>, <record>, and <transfer>) generally may contain the following elements:

• <filled> elements containing some action to execute at the moment the result field is filled in.

• <property> elements to specify properties that are in effect for this field item.

• <prompt> elements to specify prompts to be played when this field is visited.

• <grammar> elements to specify allowable spoken and character input for this field item.

• <catch> elements and catch shorthands that are in effect for this field item.

Each field item may have an associated set of shadow variables. Shadow variables are used to return results from the execution of a field item, other than the value stored under the name attribute. For example, it may be useful to know the confidence level that was obtained as a result of a recognized grammar in a <field> element. A shadow variable is referenced as name$.shadowvar where name is the value of the field item’s name attribute, and shadowvar is the name of a specific shadow variable. For example, the <field> element returns a shadow variable confidence. The code fragment below illustrates how this shadow variable is accessed.

<field name="state">
  <prompt> Please say the name of a state. </prompt>
  <grammar src="http://mygrammars.example.com/states.grm" 
              type="application/grammar"/>
  <filled>
    <if cond="state$.confidence &lt; 0.4">
      <throw event="nomatch"/>
    </if>
  </filled>
</field>

In the example, the confidence of the result is examined, and the result is rejected if the confidence is too low.

2.3.1 FIELD

A field specifies an input item to be gathered from the user. Attributes of fields include:

The shadow variables of a <field> element whose name is name are the same used in the application.lastresult$ array, name$.confidence, name$.utterance, name$.inputmode, and name$.interpretation. The value of each of these shadow variables will necessarily be the same as that found in first element of the array: application.lastresult$[0].confidence, application.lastresult$[0].utterance, application.lastresult$[0].inputmode, and application.lastresult$[0].interpretation, respectively. See Section 5.1.5 for a description of the contents of these variables.

Issues:

• The definition for precedence of grammars in Section 3.1.5 may, in certain circumstances, conflict with the requirement above that 0th entry in the n-best provide the contents of a field's shadow variables. A future working draft will resolve this conflict.

2.3.1.1 Fields Using builtin Grammars

The <field> type attribute is used to specify a builtin grammar for one of the fundamental types, and also specifies how its value is to be spoken if subsequently used in a value attribute in a prompt. An example:

<field name="lo_fat_meal" type="boolean">

  <prompt>
    Do you want a low fat meal on this flight?
  </prompt>
  <help>
    Low fat means less than 10 grams of fat, and under
    250 calories.
  </help>
  <filled>
    <prompt>
      I heard <emphasis><value expr="lo_fat_meal"/></emphasis>.
    </prompt>
  </filled>
</field>

In this example, the boolean type indicates that inputs are various forms of true and false. The value actually put into the field is either true or false. The field would be read "yes" or "no" in prompts.

In the next example, digits indicates that input will be spoken or keyed digits. The result is stored as a string, and rendered as digits, i.e., "one-two-three", not "one hundred twenty-three". The <filled> action tests the field to see if it has 12 digits. If not, the user hears the error message.

<field
name="ticket_num" type="digits">
  <prompt>
     Read the 12 digit number from your ticket.
  </prompt>
  <help>The 12 digit number is to the lower left.</help>
  <filled>
     <if cond="ticket_num.length != 12">
       <prompt>
         Sorry, I didn't hear exactly 12 digits.
       </prompt>
       <assign name="ticket_num" expr="undefined"/>
     </if>
  </filled>
</field>

Each builtin type has a convention for the format of the value returned. These are independent of locale and of the implementation. The return type for builtin fields is string except for the boolean field type. To access the actual recognition result, the author can reference the shadow variable name$.utterance.

The builtin types are defined in such a way that a VoiceXML application developer can assume some consistency of user input across implementations. This permits help messages and other prompts to be independent of platform in many instances. For example, the boolean type’s grammar should minimally allow "yes" and "no" responses, but each implementation is free to add other choices, such as "yeah" and "nope".

In cases where an application requires specific behavior or different behavior than defined for a builtin, it should use an explicit field grammar. The following are circumstances in which an application must provide an explicit field grammar in order to ensure portability of the application with a consistent user interface

A platform is not required to implement a grammar that accepts all possible values that might be returned by a builtin. For instance, the currency builtin defines the return value formatting for a very broad range of currencies (by reference to ISO 4217:1995). The platform is not required to support spoken input that includes any of the world's currencies since that can negatively impact recognition accuracy. Similarly, the number builtin can return positive or negative floating point numbers but the grammar is not required to support all possible spoken floating point numbers.

Builtins are also limited in their ability to handle underspecified spoken input. For instance, "20 peso" cannot be resolved to a specific ISO 4217:1995 currency code because the "peso" is the name of the currency of numerous nations. In such cases the platform may return a specific currency code according to the locale or may omit the currency code.

All builtin types must support both voice and DTMF entry.

The builtin types are:

2.3.1.2. Fields Using Explicit Grammars

Explicit grammars can be specified via a URI, which can be absolute or relative:

<field name="flavor">
  <prompt>What is your favorite ice cream?</prompt>
  <grammar src="../grammars/ice_cream.grxml"
           type="application/grammar+xml"/>
</field>

Grammars can be specified inline, for example using a W3C Augmented BNF grammar:

<field name="flavor">
  <prompt>What is your favorite flavor?</prompt>
  <help>Say one of vanilla, chocolate, or strawberry.</help>
  <grammar type="application/grammar">
   vanilla | chocolate | strawberry
  </grammar>
</field>

If both the src attribute and an inline grammar are provided the grammar identified by the src attribute takes precedence.

2.3.1.3. Fields Using Option Lists

When a simple set of alternatives is all that is needed to specify the legal input values for a field, it may be more convenient to use an option list than a grammar. An option list is represented by a set of <option> elements contained in a <field> element. Each <option> element contains PCDATA that is used to generate a grammar for the spoken input it accepts using the same method described for <choice>. It also has attributes specifying the DTMF key for selecting the option and the value to assign to the field when the option is chosen.

The following field offers the user three choices and assigns the value of the value attribute of the selected option to the maincourse variable:

<form>
  <field name="maincourse">
    <prompt>
      Please select an entree. Today, we’re featuring <enumerate/>
    </prompt>

    <option dtmf="1" value="fish"> swordfish </option>
    <option dtmf="2" value="beef"> roast beef </option>
    <option dtmf="3" value="chicken"> frog legs </option>

    <filled>
      <submit next="/cgi-bin/maincourse.cgi"
        method="post" namelist="maincourse"/>
    </filled>
  </field>
</form>

This conversation might sound like:

C: Please select an entree. Today, we’re featuring swordfish; roast beef; frog legs.

H: frog legs

C: (assigns "chicken" to "maincourse", then submits "maincourse=chicken" to /maincourse.cgi)

The following example shows proper and improper use of <enumerate> in a catch element of a form with several fields containing <option> elements:

<form>
  <block>
    We need a few more details to complete your order.
  </block>
  <field name="color">
    <prompt>Which color?</prompt>
    <option>red</option>
    <option>blue</option>
    <option>green</option>
  </field>
  <field name="size">
    <prompt>Which size?</prompt>
    <option>small</option>
    <option>medium</option>
    <option>large</option>
  </field>
  <field name="quantity" type="number">
    <prompt>How many?</prompt>
  </field>
  <block>
    Thank you.  Your order is being processed.
    <submit next="details.cgi"/>
  </block>
  <catch event="help nomatch">
    Your options are <enumerate/>.
  </catch>
</form>

A scenario might be:

C: We need a few more details to complete your order. Which color?

H: help. (throws "help" event caught by form-level <catch>)

C: Your options are red, blue, green.

H: red.

C: Which size?

H: 7 (throws "nomatch" event caught by form-level <catch>)

C: Your options are small, medium, large.

H: small.

In the steps above, the <enumerate/> in the form-level catch had something to enumerate: the <option> elements in the "color" and "size" <field> elements. The next <field>, however, is different:

C: How many?

H: a lot. (throws "nomatch" event caught by form-level <catch>)

The form-level <catch>'s use of <enumerate> causes an "error.semantic" event to be thrown because the "quantity" <field> does not contain any <option> elements that can be enumerated.

One solution is to add a field-level <catch> to the "quantity" <field>:

<catch event="help nomatch">
       Please say the number of items to be ordered.
</catch>

The "nomatch" event would then be caught locally, resulting in the following possible completion of the scenario:

C: Please say the number of items to be ordered.

H: 50

C: Thank you. Your order is being processed.

The <enumerate> element is also discussed in Section 2.2.

The attributes of <option> are:

The use of <option> does not preclude the simultaneous use of <grammar>. The result would be the match from either 'grammar', not unlike the occurence of two <grammar> elements in the same <field> representing a disjunction of choices.

2.3.1.4. builtin Grammars

Fundamental builtin grammars are explicitly referenced using the special-purpose "builtin:" URI scheme which allows access to resources such as speech grammars, DTMF grammars and audio files. In addition, the "builtin:" URI scheme may also be used to access platform-specific builtin grammars that are supported by particular interpreter contexts. It is recommended that plaform-specific builtin grammar names begin with the string "x-", as this namespace will not be used in future versions of the standard.

Examples of fundamental builtin grammars:

<grammar src="builtin:grammar/boolean"/>
<grammar src="builtin:dtmf/boolean"/>

where the first <grammar> references the builtin boolean speech grammar, and the second references the builtin boolean DTMF grammar.

Examples of platform-specific builtin grammars:

<grammar src="builtin:grammar/x-sample"/>
<grammar src="builtin:dtmf/x-sample"/>

2.3.1.5. builtin Parameters

Some builtin field types and grammars can be parameterized. This may be done by explicitly referring to builtin grammars using the special-purpose "builtin:" URI scheme and a URI-style query syntax of the form type?param=value in the src attribute of a <grammar> element, or in the type attribute of a field, for example:

<grammar src="builtin:dtmf/boolean?y=7;n=9"/>

<field type="boolean?y=7;n=9>...</field>

<field type="digits?minlength=3;maxlength=5">...</field>

Where the <grammar> parameterizes the builtin DTMF grammar, the first <field> parameterizes the builtin DTMF grammar (the speech grammar will be activated as normal) and the second <field> parameterizes both builtin DTMF and speech grammars. Parameters which are undefined for a given grammar type will be ignored; for example, "builtin:grammar/boolean?y=7".

By definition the following:

<field type="sample">...</field>

is equivalent to:

<field>
<grammar src="builtin:grammar/sample"/>
<grammar src="builtin:dtmf/sample"/> 
...
</field>

where sample is one of the fundamental builtin field types (e.g., boolean, date, etc.). The digits and boolean grammars can be parameterized as follows:

Note that more than one parameter may be specified separated by ";" as illustrated above. In <grammar> elements, the src attribute URI must start with builtin:grammar/ or builtin:dtmf/ as shown above. When a <grammar> element with the mode set to "voice" (the default value) is specified in a <field>, it overrides the default speech grammar implied by the type attribute of the field. Likewise, when a <grammar> element with the mode set to "dtmf" is specified in a <field>, it overrides the default DTMF grammar.

2.3.2 BLOCK

This element is a form item. It contains executable content that is executed if the block’s form item variable is undefined and the block's cond attribute, if any, evaluates to true.

<block>
   Welcome to Flamingo, your source for lawn ornaments.
</block>

The form item variable is automatically set to true just before the block is entered. Therefore, blocks are typically executed just once per form invocation.

Sometimes you may need more control over blocks. To do this, you can name the form item variable, and set or clear it to control execution of the <block>. This variable is declared in the dialog scope of the form.

Attributes of <block> include:

2.3.3. INITIAL

In a typical mixed initiative form, the <initial> element is visited when the user is initially being prompted for form-wide information, and has not yet entered into the directed mode where each field is solicited individually. Like field items, it has prompts, catches, and event counters. Unlike field items, <initial> has no grammars, and no <filled> action. For instance:

<form id="get_from_and_to_cities">
<grammar src="http://www.directions.example.com/grammars/from_to.grxml" 
       type="application/grammar+xml"/>
   <block>
       Welcome to the Driving Directions By Phone.
   </block>
   <initial name="bypass_init">
     <prompt>
       Where do you want to drive from and to?
     </prompt>
     <nomatch count="1">
        Please say something like "from Atlanta Georgia to Toledo Ohio".
     </nomatch>
     <nomatch count="2">
        I'm sorry, I still don't understand.
        I'll ask you for information one piece at a time.
        <assign name="bypass_init" expr="true"/>
        <reprompt/>
     </nomatch>
   </initial>
   <field name="from_city">
     <grammar src="http://www.directions.example.com/grammars/city.grxml" 
              type="application/grammar+xml"/>
     <prompt>From which city are you leaving?</prompt>
       ... etc. ...
   </field>
   ... etc. ...
</form>

While visiting an <initial> element, no field grammar is active. If an event occurs while visiting an <initial>, then one of its event handlers executes. As with other form items, <initial> continues to be eligible to be visited while its form item variable is undefined and while its cond attribute is true. If one or more of the field item variables is set by user input, then all <initial> form item variables are set to true, before any <filled> actions are executed.

An <initial> form item variable can be manipulated explicitly to disable, or re-enable the <initial>'s eligibility to the FIA. For example, in the program above, the <initial>'s form item variable is set on the second nomatch event. This causes the FIA to no longer consider the <initial> and to choose the next form item, which is a <field> to prompt explicitly for the origination city. Similarly, an <initial>’s form item variable could be cleared, so that <initial> gets selected again by the FIA.

Note: explicit assignment of values to field item variables does not affect the value of an <initial>’s form item variable.

Attributes of <initial> include:

2.3.4 SUBDIALOG

A <subdialog> element invokes a 'called' dialog (known as the subdialog) identified by its src attribute in the 'calling' dialog. The subdialog executes in a new execution context that includes all the declarations and state information for the dialog, the dialog’s document, and the application root (if present), with counters reset, and variables initialized. The subdialog proceeds until the execution of a <return> element which causes the subdialog to return. When the subdialog returns, its execution context is deleted, and execution resumes in the calling dialog with any appropriate <filled> elements.

The subdialog context and the context of the callee dialog are independent, even if the dialogs are in the same document. Variables in the scope chain of the calling dialog are not shared with the called subdialog: there is no sharing of variable instances between execution contexts. Even when the subdialog is specified in the same document as the calling dialog, its execution context contains different variable instances. When the subdialog and calling dialog are in different documents but share a root document, the subdialog's root variables are likewise different instances. All variable bindings applied in the subdialog context are lost on return to the calling context.

Within the subdialog context, however, normal scoping rules for grammars, events and variables apply. Active grammars in a subdialog include default grammars defined by the interpreter context and appropriately scoped grammars in <link>, <menu> and <form> elements in the subdialog's document and its root document. Event handling and variable binding likewise follow the standard scoping hierarchy.

From a programming perspective, subdialogs behave differently from subroutines because the calling and called contexts are independent. While a subroutine can access variable instances in its calling routine, a subdialog cannot access the same variable instance defined in its calling dialog. Similarly, subdialogs do not follow the event percolation model in languages like Java where an event thrown in a subrountine automatically percolates up to the calling context if not handled in the called context. Events thrown in a subdialog are treated by event handlers defined within its context; they can only be passed to the calling context by a local event handler which explicitly returns the event to the calling context (see Section 5.3.10).

Subdialogs can permit the reuse of a common dialog and can be used to build libraries of reusable applications.

The attributes are:

The <subdialog> element may contain elements common to all form items, and may also contain <param> elements. The <param> elements of a <subdialog> specify the parameters to pass to the subdialog. These parameters must be declared in the subdialog using <var> elements; it is a semantic error to attempt to set a form item variable or an undeclared variable using <param>. When a subdialog initializes, its variables are initialized in document order to the corresponding <param> value. An expr attribute in the <var> element is ignored in this case. If no corresponding <param> is specified to <var> element, an expr attribute is used as a default value, or the variable is undefined if the expr attribute is unspecified as with the regular <form> element.

In the example below, the birthday of an individual is used to validate their driver's license. The src attribute of the subdialog refers to a form that is within the same document. The <param> element is used to pass the birthday value to the subdialog.

<!-- form dialog that calls a subdialog -->
<form>
  <subdialog name="result" src="#getdriverslicense">
   <param name="birthday" expr="'2000-02-10'"/>
   <filled>
     <submit next="http://myservice.example.com/cgi-bin/process"/>
   </filled>
  </subdialog>
</form>

<!-- subdialog to get drivers license -->
<form id="getdriverslicense">
  <var name="birthday"/>
  <field name="drivelicense">
   <grammar src="http://grammarlib/drivegrammar.grxml"
      type="application/grammar+xml"/>
   <prompt> Please say your driver's license. </prompt>
   <filled>
     <if cond="validdrivelicense(drivelicense,birthday)">
       <var name="status" expr="true"/>
     <else/>
       <var name="status" expr="false"/>
     </if>
     <return namelist="drivelicense status"/>
   </filled>
  </field>
</form>

The driver’s license value is returned to calling dialog, along with a status variable in order to indicate whether the license is valid or not.

This example also illustrates the convenience of using <param> as a means for forwarding data to the subdialog as a means of instantiating values in the subdialog without using server side scripting. An alternate solution that uses scripting, is shown below.

Document with form that calls a subdialog

<?xml
version="1.0"?>
<vxml version="2.0">

  <form>
   <field name="birthday" type="date">
     What is your birthday?
   </field>
   <subdialog name="result"
          src="/cgi-bin/getlib#getdriverslicense"
          namelist="birthday">
     <filled>
       <submit next="http://myservice.example.com/cgi-bin/process"/>
     </filled>
   </subdialog>
  </form>
</vxml>

Document containing the subdialog (generated by /cgi-bin/getlib)

<?xml version="1.0"?>
<vxml version="2.0">

  <form id="getdriverlicense">
    <var name="birthday" expr="'1980-02-10'"/>
    <!-- Generated by server script -->
   <field name="drivelicense">
     <grammar src="http://grammarlib/drivegrammar.grxml"
        type="application/grammar+xml"/>
     <prompt>
       Please say your driver’s license number.
     </prompt>
     <filled>
       <if cond="validdrivelicense(drivelicense,birthday)">
         <var name="status" expr="true"/>
       <else/>
         <var name="status" expr="false"/>
       </if>
       <return namelist="drivelicense status"/>
     </filled>
   </field>
  </form>
</vxml>

In the above example, a server side script had to generate the document and embed the birthday value.

One last example is shown below that illustrates a subdialog to capture general credit card information. First the subdialog is defined in a separate document; it is intended to be reusable across different applications. It returns a status, the credit card number, and the expiry date; if a result cannot be obtained, the status is returned with value "no_result".

<?xml version="1.0"?>
<vxml version="2.0">
   <!-- Example of subdialog to collect credit card information. -->
   <!-- file is at http://www.somedomain.example.com/ccn.vxml -->
   <form id="getcredit">
     <var name="status" expr="'no_result'"/>

     <field name="creditcardnum">
       <prompt>
         What is your credit card number?
       </prompt>
       <help>
         I am trying to collect your credit card information.
         <reprompt/>
       </help>
       <nomatch>
         <return namelist="status"/>
       </nomatch>
       <grammar src="ccn.grxml" type="application/grammar+xml"/>
     </field>

     <field name="expirydate" type="date">
       <prompt>
         What is the expiry date of this card?
       </prompt>
       <help>
         I am trying to collect the expiry date of the credit
         card number you provided.
         <reprompt/>
       </help>
       <nomatch>
         <return namelist="status"/>
       </nomatch>
     </field>

     <block>
       <assign name="status" expr="'result'"/>
       <return namelist="status creditcardnum expirydate"/>
     </block>
   </form>
</vxml>

An application that includes a calling dialog is shown below. It obtains the name of a software product and operating system using a mixed initiative dialog, and then solicits credit card information using the subdialog.

<?xml version="1.0"?>
<vxml version="2.0">
  <!-- Example main program -->
  <!-- http://www.somedomain.example.com/main.vxml -->
  <!-- calls subdialog ccn.vxml -->

  <!-- assume this gets defined by some dialog -->
  <var name="username"/>

  <form id="buysoftware">
    <var name="ccn"/>
    <var name="exp"/>
    <grammar src="buysoftware.grxml" type="application/grammar+xml"/>

    <initial name="start">
      <prompt>
        Please tell us the software product you wish to buy
        and the operating system on which it must run.
      </prompt>
      <noinput>
        <assign name="start" expr="true"/>
      </noinput>
    </initial>

    <field name="product">
      <prompt>
        Which software product would you like to buy?
      </prompt>
    </field>

    <field name="operatingsystem">
      <prompt>
        Which operating system does this software need to run on?
      </prompt>
    </field>

    <subdialog name="cc_results"
        src="http://somedomain.example.com/ccn.vxml">
      <filled>
        <if cond="cc_results.status=='no_result'">
          Sorry, your credit card information could not be
          Obtained. This order is cancelled.
          <exit/>
        <else/>
          <assign name="ccn" expr="cc_results.creditcardnum"/>
          <assign name="exp" expr="cc_results.expirydate"/>
        </if>
      </filled>
    </subdialog>

    <block>
      We will now process your order. Please hold.
      <submit next="www.somedomain.example.com/process_order.asp"
              namelist="username product operatingsystem ccn exp"/>
    </block>
  </form>
</vxml>

2.3.5. OBJECT

A VoiceXML implementation platform may expose platform-specific functionality for use by a VoiceXML application via the <object> element. The <object> element makes direct use of its own content during initialization (e.g. <param> child element) and execution. As a result, <object> content cannot be treated as alternative content.

For example, a platform-specifc credit card collection object could be accessed like this:

<object
    name="debit"
    classid="method://credit_card/gather_and_debit"
    data="http://www.recordings.example.com/prompts/credit/jesse.jar">
  <param name="amount" expr="document.amt"/>
  <param name="vendor" expr="vendor_num"/>
</object>

In this example, the <param> element (Section 6.4) is used to pass parameters to the object when it is invoked. When this <object> is executed, it returns an ECMAScript object as the value of its form item variable. This <block> presents the values returned from the credit card object:

<block>
   <prompt>
     The card type is <value expr="debit.card"/>.
   </prompt>

   <prompt>
     The card number is <value expr="debit.card_no"/>.
   </prompt>

   <prompt>
     The expiration date is <value expr="debit.expiry_date"/>.
   </prompt>

   <prompt>
     The approval code is <value expr="debit.approval_code"/>.
   </prompt>

   <prompt>The confirmation number is
     <value expr="debit.conf_no"/>.
   </prompt>
</block>

As another example, suppose that a platform has a feature that allows the user to enter arbitrary text messages using a telephone keypad.

<form id="gather_pager_message">
  <object name="message"
      classid="builtin://keypad_text_input">
   <prompt>
     Enter your message by pressing your keypad once
     per letter.  For a space, enter star.  To end the
     message, press the pound sign.
   </prompt>
  </object>

  <block>
    <assign name="document.pager_message" expr="message.text"/>
    <goto next="#confirm_pager_message"/>
  </block>
</form>

The user is first prompted for the pager message, then keys it in. The <block> copies the message to the variable document.pager_message.

Attributes of <object> include:

If an <object> element refers to an unknown object, the error.unsupported.object event is thrown. There is no requirement for implementations to provide platform-specific objects, although implementations must handle the <object> element by throwing error.unsupported.object if the particular platform-specific object is not supported.

2.3.6. RECORD

The <record> element is a field item that collects a recording from the user. A reference to the recorded audio is stored in the field item variable, which can be played back (using the expr attribute on either <value> or <audio>) or submitted to a server, as shown in this example:

<?xml version="1.0"?>
<vxml version="2.0">
   <form>
     <record  name="greeting" beep="true" maxtime="10s"
        finalsilence="4000ms" dtmfterm="true" type="audio/wav">
       <prompt>
         At the tone, please say your greeting.
       </prompt>
       <noinput>
         I didn't hear anything, please try again.
       </noinput>
     </record>

     <field name="confirm" type="boolean">
       <prompt>
         Your greeting is <value expr="greeting"/>.
       </prompt>
       <prompt>
         To keep it, say yes.  To discard it, say no.
       </prompt>
       <filled>
         <if cond="confirm">
           <submit next="save_greeting.pl"
              method="post" namelist="greeting"/>
         </if>
         <clear/>
       </filled>
     </field>
   </form>
</vxml>

The user is prompted for a greeting and then records it. The recording terminates when one of the following conditions is met: the interval of final silence occurs, a DTMF key is pressed, the maximum recording time is exceeded, or the caller hangs up. The greeting is played back, and if the user approves it, is sent on to the server for storage using the HTTP POST method. Notice that like other field items, <record> has prompts and catch elements. It may also have <filled> actions.

When a user hangs up during recording, the recording terminates and a telephone.disconnect.hangup event is thrown. However, audio recorded up until the hangup is available through the <record> variable. Applications, such as simple voicemail services, can then return audio data to a server even afer disconnection:

<form>
   <record name="voicemail">
   ...
      <catch name="telephone.disconnect.hangup">
           <submit next="./voicemail_server.asp"/>
      </catch>
   </record>
   ...
</form>

If the platform supports simultaneous recognition and recording, then the local grammar may be used for termination of the recording. The 'terminating' speech input is accessible via the "utterance" shadow variable; its confidence via the "confidence" shadow variable. The audio of the recognized 'terminating' speech input is not available and is not part of the recording (as with DTMF termination). Form and document scoped grammars may also be active while the recording is in progress. If a form or document scoped grammar is matched, recording is terminated, the record element is left unfilled, and the form interpretation algorithm is invoked.

The attributes of <record> are:

The <record> shadow variable name$ has the following ECMAScript properties after the recording has been made:

name$.duration

The duration of the recording in milliseconds.

name$.size

The size of the recording in bytes.

name$.termchar

If the dtmfterm attribute is true, and the user terminates the recording by pressing a DTMF key, then this shadow variable is the key pressed (e.g. "#"). Otherwise it is null.

name$.maxtime

Boolean, true if the recording was terminated because the maxtime duration was reached.

name$.dest

URI referencing the recording. If the destination redirects the original request, for instance in the case of RTSP or HTTP servers, this variable holds the redirected URI.

Issues:

• The mime type of the recording is not always sufficiently specific to identify the precise audio encoding format; for example, the "audio/wav" mime type does not identify the codec used. RFC 2361 defines a method of referencing CODEC IDs in the Microsoft WAVE registry, e.g. "audio/vnd.wave; codec=123". A future revision of this document may include an attribute which specifies the precise encoding.

2.3.7. TRANSFER

Occasionally, it is appropriate to suspend the session between the user and the interpreter and initiate a session with another entity. The most common use for this capability in current practice is to connect a user in a telephone conversation with an interpreter to a third party through the telephone network. The <transfer> element directs the interpreter to make such a third party connection. Two scenarios are supported:

The form item variable is used to store the outcome of the transfer attempt. In the case of a blind transfer, the value of the form item variable is undefined. The possible values for bridge transfers are:

The following example attempts to transfer the user to a customer support operator and then wait for that conversation to terminate. An audio file is played before actually performing the transfer; this is useful in cases where the it is desired to inform the caller of what is happening, with a notice such as "Please wait while we transfer your call". After the audio play completes, the transfer begins. The caller is connected to the outgoing telephony channel. They will hear the audio from the far end (including call progress tones such as ringing, busy, network busy, etc.) The "transferaudio" attribute specifies an audio file to be played to the caller in place of audio from the far end until the far end answers. The audio will stop playing immediately upon far-end answer. If the audio file ends before the far-end answers, the caller will be connected to the far-end audio (thus hearing call progress tones).

<form id="xfer">
   <var name="mydur" expr="0"/>
   <block>
   <prompt>
   <!-- queue and played before starting the transfer -->
   <!-- "Please wait while we transfer your call..." -->
       <audio src="please_wait.wav">
   </prompt>
   </block>

  <!-- play elevator music while ringing;
       wait up to 60 seconds for the far end to answer  -->
   <transfer name="mycall" dest="tel:+358-555-1234567"
      transferaudio="elevator_music.wav"
      connecttimeout="60s" bridge="true">

     <filled>
       <assign name="mydur" expr="mycall$.duration"/>
         <if cond="mycall == 'busy'">
           <prompt>
             Sorry, our customer support team is busy serving
             other customers.  Please try again later.
           </prompt>
         <elseif cond="mycall == 'noanswer'"/>
           <prompt>
             Sorry, our customer support team's normal hours
             are 9 am to 7 pm Monday through Saturday.
           </prompt>
         </if>
     </filled>
   </transfer>
   <block>
      <submit namelist="mycall mydur" next="/cgi-bin/report"/>
   </block>
</form>

During a bridge transfer, the platform can listen for DTMF and voice input from the caller. In particular, if a DTMF grammar appears inside the <transfer> element, DTMF input matching that grammar will terminate the transfer and return control to the interpreter. Similarly, a bridge transfer may be terminated by recognition of an utterance matching an enclosed <grammar> element; support for speech recognition during transfer is not required. The <transfer> element is modal in that no grammar defined outside its scope is active. When transfer is terminated by DTMF or speech, a near_end_disconnect value is returned.

As is the case with other field items, <prompt> elements within <transfer> are queued and played before the transfer begins.

A <grammar> included within a <transfer> is active during playing of prompts, and during the transfer.

The <transfer> element is optional, though platforms should support it.

Attributes include:

The <transfer> shadow variable (name$) has the following ECMAScript properties after a transfer completes:

name$.duration

The duration of a successful call in seconds (floating-point). The duration is 0 if the transferred call was terminated prior to being answered.

Events thrown inside a <transfer> include:

telephone.disconnect.hangup

If the caller hung up.

telephone.disconnect.transfer

If the caller has been transferred unconditionally to another line and will not return.

error.telephone.noauthorization

If the caller is not allowed to call the destination.

error.telephone.baddestination

If the destination URI is malformed.

error.telephone.noroute

If the platform is not able to place a call to the destination.

error.telephone.noresource

If the platform cannot allocate resources to place the call.

error.telephone.protocol.nnn

The telephony protocol stack raised an exception that does not correspond to one of the other error.telephone events.

2.4 Filled

The <filled> element specifies an action to perform when some combination of fields are filled by user input. It may occur in two places: as a child of the <form> element, or as a child of a field item.

As a child of a <form> element, the <filled> element can be used to perform actions that occur when a combination of one or more fields is filled. For example, the following <filled> element does a cross-check to ensure that a starting city field differs from the ending city field:

<form id="get_starting_and_ending_cities">
  <field name="start_city">
      <grammar src="http://www.grammars.example.com/voicexml/city.grxml" 
        type="application/grammar+xml"/>
      <prompt>What is the starting city?</prompt>
  </field>
  <field name="end_city">
      <grammar src="http://www.grammars.example.com/voicexml/city.grxml" 
        type="application/grammar+xml"/>
      <prompt>What is the ending city?</prompt>
  </field>
  <filled mode="any" namelist="start_city end_city">
    <if cond="start_city == end_city">
      <prompt>
        You can't fly from and to the same city.
      </prompt>
      <clear/>
    </if>
  </filled>
</form>

If the <filled> element appears inside a field item, it specifies an action to perform after that field is filled in by user input:

<form id="get_city">
  <field name="city">
    <grammar type="application/grammar+xml"/>
      src="http://www.ship-it.example.com/grammars/served_cities.grxml"
    <prompt>What is the city?</prompt>
    <filled>
      <if cond="city == 'Novosibirsk'">
        <prompt>
          Note, Novosibirsk service ends next year.
        </prompt>
      </if>
    </filled>
  </field>
</form>

After each gathering of the user’s input, all the fields mentioned in the input are set, and then the interpreter looks at each <filled> element in document order (no preference is given to ones in fields vs. ones in the form). Those whose conditions are matched by the utterance are then executed in order, until there are no more, or until one transfers control or throws an event.

Attributes include:

2.5 Links

A <link> element may have one or more grammars, which are scoped to the element containing the <link>. Grammar elements contained in the <link> are not permitted to specify scope. When one of these grammars is matched, the link activates, and either:

• Transitions to a new document or dialog (like <goto>), or

• Throws an event (like <throw>).

For instance, this link activates when you say "books" or press "2".

<link next="http://www.voicexml.org/books/main.vxml">
  <grammar mode="voice">
    <rule id="root" scope="public">
       <one-of>
         <item>books</item> 
         <item>VoiceXML books</item> 
       </one-of>
    </rule>
  </grammar>
  <grammar mode="dtmf">
     <rule id="r2" scope="public"> 2 </rule>
  </grammar>
</link>

This link takes you to a dynamically determined dialog in the current document:

<link expr="'#' + document.helpstate">
  <grammar mode="voice">
     <rule id="root" scope="public"> help </rule>
  </grammar>
</link>

The <link> element can be a child of <vxml>, <form>, or of a form item. A link at the <vxml> level has grammars that are active throughout the document. A link at the <form> level has grammars active while the user is in that form. If an application root document has a document-level link, its grammars are active no matter what document of the application is being executed.

If execution is in a modal form item, then link grammars at the form or document level are not active.

You can also define a link that, when matched, throws an event instead of going to a new document. This event is thrown at the current location in the execution, not at the location where the link is specified. For example, if the user matches this link’s grammar or enters '2' on the keypad, a help event is thrown in the form item the user was visiting and is handled by the best qualified <catch> in the item's scope (see Catch Element Selection for further details):

<link dtmf="2" event="help">
  <grammar mode="voice">
    <rule id="r5" scope="public">
       <one-of>
         <item>arrgh</item> 
         <item>alas all is lost</item> 
         <item>fie ye froward machine</item> 
         <item>I don't get it</item> 
       </one-of>
    </rule>
  </grammar>
</link>

Attributes of <link> are:

3. User Input

3.1 Grammars

3.1.1 Speech Grammars

The <grammar> element is used to provide a speech grammar that

• specifies a set of utterances that a user may speak to perform an action or supply information, and

• provides a corresponding string value (in the case of a field grammar) or set of attribute-value pairs (in the case of a form grammar) to describe the information or action.

The <grammar> element is designed to accommodate any grammar format that meets these two requirements. VoiceXML platforms must support at least one common format, the XML Form of the W3C Speech Recognition Grammar Format. VoiceXML platforms may support the Augmented BNF Form of the W3C Speech Recognition Grammar Format. VoiceXML platforms may support other grammar formats such as the Java Speech API Grammar Format (the format used throughout the VoiceXML 1.0 specification) or proprietary formats.

VoiceXML platforms should be a Conforming XML Grammar Processor as defined in the W3C Speech Recognition Grammar Format. While this requires a platform to process documents with one or more "xml:lang" attributes defined, it does not require that the platform must be multi-lingual. When an unsupported language, or language list, is encountered, the platform throws an error.unsupported.language event which specifies the unsupported language(s) in its message variable.

The following elements are defined for grammar support. This document does not redefine these elements. Refer to the W3C Speech Recognition Grammar Format for definitions and examples.

Isssues:

• It is intended that grammar element described in this document is synchronized with the W3C Speech Recognition Grammar Format specification. Where there are discrepanies during the W3C process, Speech Recognition Grammar Format definitions take precedence.
• A future version of this document use a schema approach where the grammar element is specified by reference to a separate namespace defined in the W3C Speech Recognition Grammar Format document.

3.1.1.1 Inline Grammars

The <grammar> element may be used to specify an inline grammar or an external grammar. An inline grammar is specified by the content of a <grammar> element and defines an entire grammar:

<grammar type="mime-type" mode="voice">
     inline speech grammar
</grammar>

It may be necessary in this case to enclose the content in a CDATA section. For inline grammars the type parameter specifies a media type that governs the interpretation of the content of the <grammar> tag.

The following is an example of grammar defined by the XML Form of the W3C Speech Recognition Grammar Format.

<grammar mode="voice" xml:lang="en-US">
  <!-- Command is an action on an object -->
  <!-- e.g. "open a window" -->
  <rule id="command" scope="public">
    <ruleref uri="#action"/> <ruleref uri="#object"/>
  </rule>

  <rule id="action">
    <one-of>
      <item> open </item>
      <item> close </item>
      <item> delete </item>
      <item> move </item>
    </one-of>
  </rule>

  <rule id="object">
   <count number="optional">
      <one-of> <item> the </item> <item> a </item> </one-of>
    </count>
    <one-of>
      <item> window </item>
      <item> file </item>
      <item> menu </item>
    </one-of>
  </rule>
</grammar>

The following is the equivalent example of grammar defined by the Augmented BNF Form of the W3C Speech Recognition Grammar Format. Because VoiceXML platforms are not required to support this format it may be less portable.

<grammar mode="voice" type="application/grammar">
  public $command = $action $object;
  $action = open | close | delete | move;
  $object = [the | a] (window | file | menu);
</grammar>

Issues

• The capabilities for inline grammars in this working draft are consistent with the spirit of included grammars in VoiceXML 1.0. Both XML and non-XML grammar formats (e.g. the W3C ABNF format or JSGF) can be included. This represents an overloading of the grammar element and requires the use of mixed-content in the DTD. This permits incompatibilities between the W3C XML Grammar format and the XML grammars permitted within VoiceXML 2.0. We are considering addressing the issue in a future working draft by an approach such as
  1. Strict mode: permit only XML grammars to be included within VoiceXML 2.0
  2. Loose: as in the current draft allow the grammar element to include a text grammar as CDATA or include an XML grammar with a looser DTD than in the grammar spec
  3. Nested: introduce a new element to contain textual (non-XML) grammars such as W3C ABNF or JSGF.
  (See also the equivalent issues for grammar fragments in the next section.)

3.1.1.2 Inline Grammar Fragments

The <grammar> element may also be used to specify an inline grammar fragment. A grammar fragment is defined as a legal rule expansion in the grammar format. (Note: the W3C grammar specification does not yet address conformance behavior for grammar fragments.)

Note: the specification does not currently permit inline grammar fragments to be defined with the XML grammar format. This issue is under study (see the Issues below).

<grammar type="mime-type" mode="voice">
     inline speech grammar fragment
</grammar>

Grammar fragments are a useful short-hand when the inline grammar represents a short phrase or set of short phrases.

It may be necessary in this case to enclose the content in a CDATA section. For inline grammars the type parameter specifies a media type that governs the interpretation of the content of the <grammar> tag.

The following is an example of a grammar fragment defined with the Augmented BNF Form of the W3C Speech Recognition Grammar Format. Because VoiceXML platforms are not required to support this format it might be less portable.

<grammar mode="voice" type="application/grammar">
  red | green | blue
</grammar>

Issues

• The current draft does not permit inline grammar fragments to be defined with the XML grammar format. The motivation is to avoid unnecessary incompatibilities with the W3C XML Grammar Format. The issue remains under consideration and may change if the grammar specification defines conformance behavior with respect to XML fragments and after the impact of permitting XML grammar fragments has been studied. In the mean time, developers should wrap XML rule expansions with a single rule element. For example, the equivalent XML form of the example above is:

<grammar>
  <rule id="root" scope="public">
    <one-of>
      <item> red </item>
      <item> green </item>
      <item> blue </item>
    </one-of>
  </rule>
</grammar>

3.1.1.3 External Grammars

An external grammar is specified by an element of the form

<grammar src="URI" type="media-type"/>

The media type is optional in this case because this information may be obtained via the URI protocol (as in the case of HTTP), and may be inferred from the filename extension. If the type is not specified, and cannot be inferred, the default type is platform specific. However, if the type is specified using the type attribute, it overrides other information about the type.

If the src attribute is defined and there is an inline grammar as content of a grammar element then the src attribute takes precedence.

The following is an example of a reference to an external grammar written in the W3C XML Grammar Format.

<grammar type="application/grammar+xml"
  src="http://www.grammar.example.com/date.grxml"/>

The following example is the equivalent grammar reference for a grammar that is authored using the W3C Augmented BNF grammar format.

<grammar type="application/grammar"
  src="http://www.grammar.example.com/date.grm"/>

The following example is the equivalent grammar reference for a grammar that is authored using the JSpeech Grammar Format. The example illustrates that VoiceXML 2.0 does not restrict which grammar formats a VoiceXML platform may support. Since the W3C XML Grammar Format is the only format that a platform is required to support it should be used when interoperability and portability are required.

<grammar type="application/x-jsgf"
  src="http://www.grammar.example.com/date.gram"/>

Note: the filename suffixes and media types in these examples are tentative.

3.1.1.4 Grammar Weight

A weight for the grammar can be specified by the weight attribute:

<grammar weight="0.6" src="form.grxmlm" type="application/grammar+xml"/>

Grammar elements, including those in link, field and form elements, can have a weight attribute. The grammar can be inline, external or built-in. The value of the weight attribute indicates the occurrence likelyhood of each grammar, and must be zero or a positive floating point number (nnn[.nnn]). The default weight for a grammar whose weight is unspecified  is 1.0. Consequently, if no weight is specified for any grammar element, all grammars are equally likely. The semantics of weight attributes is equivalent to the case where all active grammars are listed as items in alternatives (one-of element) in W3C XML Grammar Format, except for the fact that the grammar with no weight specified has default weight of 1.0.

<link event="help">
  <grammar weight="0.5" mode="voice">
   <rule id="help" scope="public">
      <count number="optional"><item>Please</item></count><item>help</item>
   </rule>  
  </grammar>
</link>

<form>
  <grammar src="form.grxml" type="application/grammar+xml"/>
  <field name="expireDate">
     <grammar weight="1.2" src="builtin:grammar/date"/>
  </field>
</form>

In the example above, the semantics of weights is equivalent to the following W3C XML grammar.

<one-of>
   <item weight="0.5"> <ruleref uri="#help"/> </item>
   <item weight="1.0"> <ruleref uri="form.grxml"/> </item>
   <item weight="1.2"> <ruleref uri="builtin:grammar/boolean"/></item>
</one-of>





<rule id="help">


  <count number="optional"><item>Please</item></count><item>help</item>


</rule>

Implicit grammars, such as those in options, do not support weights - use the <grammar> element instead for control over grammar weight.

Grammar weights only affect grammar processing. They do not directly affect the post processing of grammar results, including grammar precedence when user input matches multiple active grammar (see Section 3.1.4).

A weight has no effect on DTMF grammars (See Section 3.1.2). Any weight attribute specified  in a grammar element whose mode attribute is dtmf is ignored.

<grammar mode="dtmf" weight="0.3" src="builtin:number"/> <!-- weight will be ignored -->

Appropriate weights are difficult to determine, and guessing weights does not always improve recognition performance. Effective weights are usually obtained by study of real speech and textual data on a paricular platform. Furthermore, a grammar weight is platform specific. Note that different ASR engines may treat the same weight value differently. Therefore, the weight value that works well on particular platform may generate different results on other platforms.

3.1.1.5 Grammar Element

Attributes of <grammar> include:

When referencing an external grammar, the value of the src attribute is a URI specifying the location of the grammar with an optional fragment for the rulename. Section 2.2 of the Speech Recognition Grammar Format defines several forms of rule reference. The following are the forms that are permitted on a grammar element in VoiceXML.

• Reference to a named rule in an external grammar: src attribute is an absolute or relative URI reference to a grammar which includes a fragment with a rulename. This form of rule reference to an external grammar follows the behavior defined in Section 2.2.2 of SRGF. If the URI cannot be fetched or if the rulename is not defined in the grammar or is not a public (activatable) rule of that grammar then an error.badfetch is thrown.
• Reference to the root rule of an external grammar: src attribute is an absolute or relative URI reference to a grammar but does not include a fragment identifying a rulename. This form implicitly references the root rule of the grammar as defined in Section 2.2.2 of SRGF. If the URI cannot be fetched or if the grammar cannot be referenced by its root (see Section 4.4 of SRGF) then an error.badfetch is thrown.

The following are the forms of rule reference defined by SRGF that are not supported in VoiceXML 2.0.

• Local rule reference: a fragment-only URI is not permited. (See definition in Section 2.2.1 of SRGF). A fragment-only URI value for the src attribute causes an error.semantic event.
• Reference by alias: although an inline grammar may define aliases and make reference by alias there is no support for alias references on the grammar element itself. (See definitions in Section 2.2.3 of SRGF). There is no syntactic support for this form so no error can be generated.
• Reference to special rules: although an inline grammar may reference the special rules of SRGF (NULL, VOID, GARBAGE) there is no support for special rule references on the grammar element itself. (See definitions in Section 2.2.3 of SRGF). There is no syntactic support for this form so no error can be generated.

3.1.2 DTMF Grammars

The <grammar> element can be used to specify a DTMF grammar that

• defines a set of key presses that a user may use to perform an action or supply information, and

• defines the corresponding string value that describes that information or action.

VoiceXML platforms are required to support the DTMF grammar XML format defined in Appendix D of the W3C Grammar specification to advance application portability.

A DTMF grammar is distinguished from a speech grammar by the mode attribute on the <grammar> element. An "xml:lang" attribute has no effect on DTMF grammar handling. In other respects speech and DTMF grammars are handled identically including the ability to define the grammar inline, by an inline grammar fragment, or by an external grammar reference. The media type handling, scoping and fetching are also identical.

The following is an example of a simple XML DTMF grammar that accepts as input either "123" or "#" as input.

<grammar mode="dtmf">
  <rule id="root" scope="public">
    <one-of>
      <item> 123 </item>
      <item> # </item>
    </one-of>
  </rule>
</grammar>

3.1.3 Scope of Grammars

Field grammars are always scoped to their fields, that is, they are not active unless the interpreter is visiting that field. Grammars contained in fields cannot specify a scope.

Link grammars are given the scope of the element that contains the link. Thus, if they are defined in the application root document, links are also active in any other loaded application document. Grammars contained in links cannot specify a scope.

Form grammars are by default given dialog scope, so that they are active only when the user is in the form. If they are given scope document, they are active whenever the user is in the document. If they are given scope document and the document is the application root document, then they are also active whenever the user is in another loaded document in the same application. A grammar in a form may be given document scope either by specifying the scope attribute on the form element or by specifying the scope attribute on the <grammar> element. If both are specified, the grammar assumes the scope specified by the <grammar> element.

Menu grammars are also by default given dialog scope, and are active only when the user is in the menu. But they can be given document scope and be active throughout the document, and if their document is the application root document, also be active in any other loaded document belonging to the application. Grammars contained in menu choices cannot specify a scope.

Sometimes a form may need to have some grammars active throughout the document, and other grammars that should be active only when in the form. One reason for doing this is to minimize grammar overlap problems. To do this, each individual <grammar> element can be given its own scope if that scope should be different than the scope of the <form> element itself:

<form scope="document"> 
  <grammar> ... </grammar> 
  <grammar scope="dialog"> ... </grammar> 
</form>

3.1.4 Activation of Grammars

When the interpreter waits for input as a result of visiting a field, the following grammars are active:

• grammars for that field, including grammars contained in links in that field;

• grammars for its form, including grammars contained in links in that form;

• grammars contained in links in its document, and grammars for menus and other forms in its document which are given document scope;

• grammars contained in links in its application root document, and grammars for menus and forms in its application root document which are given document scope.

• grammars defined by platform default event handlers, such as help, exit and cancel.

In the case that an input matches more than one active grammar, the list above defines the precedence order. If the input matches more than one active grammar with the same precedence, the precedence is determined using document order. Menus behave with regard to grammar activation like their equivalent forms (see Section 2.2).

If the form item is modal (i.e., its modal attribute is set to true), all grammars except its own are turned off while waiting for input. If the input matches a grammar in a form or menu other than the current form or menu, control passes to the other form or menu. If the match causes control to leave the current form, all current form data is lost.

Issues:

• The definition above for precedence of grammars may, in certain circumstances, conflict with the requirement in Section 2.3.1 that 0th entry in the n-best provide the contents of a field's shadow variables. A future working draft will resolve this conflict.

3.1.5 Semantic Interpretation of Input

Issues:

The W3C Voice Browser Working Group is currently developing standards for processing and representation of semantic input. The following is a summary of the work in progress and the issues that are being addressed. The intent of this work is to provide a standard mechanism for each of the following:

1. Grammar annotation: A speech or DTMF grammar may be annotated with tags that indicate how to process input that matches the grammar. The current W3C Speech Recognition Grammar Specification defines the tag capabilities in Section 2.6.
2. Semantic interpretation tags: A definition of the content of those tags and of the computational process by which the tags are used to process input speech or DTMF events. This specification has not yet been published.
3. Natural Language Semantics Markup Language: A XML document that represents the semantic content of a input utterance. The first Working Draft of this specification is available.
4. VoiceXML: A representation of that semantic content in the scope of the VoiceXML Processor that permits the VoiceXML application to use the information in dialog processing.

The following is a tentative example of semantic attachments The tags represent simple keys that will be returned as the result when either the speech or DTMF grammar is matched. Whether the user says "Amex" or "American Express" or enters DTMF key '1' the result will be the "AMEX" tag.

<grammar mode="voice">
    <rule id="rule" scope="public">
      <one-of>
        <item tag="MC">master card</item>
        <item tag="VISA">visa</item>
        <item tag="AMEX">amex</item>
        <item tag="AMEX">american express</item>
      </one-of>
    </rule>
  </grammar>
  <grammar mode="dtmf">
    <rule id="rule" scope="public">
      <one-of>
        <item tag="MC"> 1 </item>
        <item tag="VISA"> 2 </item>
        <item tag="AMEX"> 3 </item>
      </one-of>
    </rule>
  </grammar>

3.1.6 Mapping Semantic Interpretation Results to VoiceXML forms

In order to make the results of the interpretation of a user's utterance usable within a VoiceXML form, the interpretation must be mapped into VoiceXML ECMAScript variables,  Results can be represented either in the proposed XML format defined by the Natural Language Semantics Markup Language  or in the ECMAScript-like output  format defined in the Speech Recognition Grammar Specification. In either case, many of the same issues arise; however, because the Natural Language Semantics Markup Language is less mature than the speech recognition output format, we focus in this discussion on examples using the ECMAScript-based ASR result format.  

This discussion deals with issues concerning mapping results from form-level and field level grammars. There is also a brief discussion of other issues such as the NL Semantics  to ECMAScript mapping, transitioning information from ASR results to VoiceXML,and dealing with mismatches between the interpretation result and the VoiceXML form.

3.1.6.1 Mapping results from form-level grammars to fields

Consider the following interpretation result from the sentence "I would like a coca cola and three large pizzas with pepperoni and mushrooms."

{    
     drink: "coke"
     pizza: {
          number: "3" 
          size: "large"
           topping: [
                "pepperoni"              
                "mushrooms"            
           ] 
      }
}

produced by the following grammar:

$order = I would like a $drink {drink:$drink} and $pizza {pizza:$pizza};

$drink = coke | pepsi | coca cola {{coke}};

$pizza = $number $size {size:$size; number:$number.n} pizzas with $tops {topping:$tops.l};

$number = (a | one){n:1} | two {n:2}| three {n:3};

$size = small | medium | large ;

$tops = $top {l:List($top)} (and $top {l:Append(l,$top)})+ ; // construct list of toppings

$top = anchovies|pepperoni|mushrooms;

The following table illustrates how the result shown above from a form-level grammar would be assigned to various fields within the form. Note that all fields that can be filled in from the interpretation are filled in simultaneously before any fields are visited by the FIA.

These examples can be explained by rules that are compatible with and are straightforward extensions of the VoiceXML 1.0 "name" and "slot" attributes:

1. The "slot" attribute of a <field> is a (very restricted) ECMAScript expression that selects some portion of the result to be assigned to the field. Note that it is possible for a specific slot value to fill more than one field, if the slot names of the fields are the same.
2. If the portion of the result named by the "slot" (or "name") attribute of a <field> doesn't exist in a given result then the field item's value is unchanged.
3. The default value for the "slot" attribute is supplied by the value of "name" attribute.

The examples only show only one level of sub-element, but it seems reasonable to allow sub-sub-elements to arbitrary levels of nesting, using a dot-separated list of element/property names. Another potential extension would be to use the slot name to select an array element, for example "slot="pizza.topping[0]". However, it's not clear how useful it would be since there's no mechanism in VoiceXML for looping through the array elements. In addition, this raises other questions, such as whether the index of the array can be a variable.

3.1.6.2 Mapping results from field-level grammars

Field-level grammars, unlike form-level grammars, are only active while visiting the field in which they are contained, and therefore they are only able to supply a value for that field. This is useful for example in directed dialogs, where a user is prompted individually for a value for each field. This is supported by assigning to the field variable the entire semantic result from a field-level grammar.

Consider for example the following field that contains a field-level grammar that returns a value representing the credit card type. (Note that the exact format of the XML grammar tags hasn't yet been determined; the example will be reconciled with this format when it is fully defined.)

<field name="cardname"/>
  <prompt>Which credit card?</prompt>
  <grammar mode="voice">
    <rule id="rule" scope="public">
      <one-of>
        <item tag="MC">master card</item>
        <item tag="VISA">visa</item>
        <item tag="AMEX">amex</item>
        <item tag="AMEX">american express</item>
      </one-of>
    </rule>
  </grammar>

When the user responds with a credit card name such as "american express" the following semantic result value might be returned:

            "AMEX"

This ECMAScript string value "AMEX" would be assigned to the "cardname" field that contains the grammar.

This assignment of the entire semantic result to a field would also apply to a structured object. An example of such a structured object might be a date, when the application treats the date as a unit.

Issues:

• For the purposes of symmetry with the form-level case, it might seem reasonable to support assignment of semantic results to slots in field-level grammars when a slot is present, in a similar fashion to the example in line 2 of the table above. However, this would preclude the useful case where the form has both a form-level grammar, using the slot attribute of a field to select a slot from the form-level grammar, and a field level grammar using simple tags to directly provide a field value ignoring the slot parameter.

3.1.6.3 Additional issues

1. Mapping from NL semantics to ECMAScript: If the NL Semantics Markup Language (NLSML) is used, a mapping needs to be defined from the NLSML representation to ECMAScript objects. Since both types of representation have similar nested structures, this mapping is fairly straightforward. This mapping is discussed in detail in the NL Semantics specification.

2. Transitioning semantic results from ASR to VoiceXML: The result of processing the semantic tags of a W3C ASR grammar is the value of the attribute of the root rule when all semantic attachment evaluations have been completed. In addition, the root rule (like all non-terminals) has an associated "text" variable which contains the series of tokens in the utterance that is governed by that non-terminal. In the process of making ASR results available to VoiceXML documents, the VoiceXML platform is not only responsible for filling in the VoiceXML fields based on the value of the attribute of the root rule, as described above, but also for filling in the shadow variables of the field. The name$.utterance shadow variable of the field should be the same as the "text" variable value for the ASR root rule. The platform is also responsible for instantiating the value of the shadow variable "name$.confidence" based on information supplied by the ASR platform, as well as the value of "name$.inputmode" based on whether DTMF or speech was processed. Finally, the platform is responsible for making this same information available in the "application.lastresult$" variable, defined in Section 5.1.5 (specifically, "application.lastresult$.utterance", "application.lastresult$.inputmode", "application.lastresult$.interpretation", and "application.lastresult$.confidence").

3. Mismatches between semantic results and VoiceXML fields: Mapping semantic results to VoiceXML depends on a tight coordination between the ASR grammar and the VoiceXML markup. Since in the current framework there's nothing that enforces consistency between a grammar and the associated VoiceXML dialog, mismatches can occur due to developer oversight. Since the dialog's behaviour during these mismatches is difficult to distinguish from certain normal situations, verifying consistency of information is extremely important. Some examples of mismatches:

• The semantic results contain extra information that doesn't correspond to the VoiceXML fields.This could occur either due to developer error or if a richer grammar is being used than is required by the VoiceXML application. This extra information will be ignored.
• The VoiceXML application expects information in the result that isn't present. This could also be due to developer error, or it may be that the user simply didn't supply a value for a particular slot. In this case the normal FIA applies and the missing value would be elicited from the user. If the problem was in fact caused by a developer error, and the grammar is actually incapable of recognizing the correct value, the FIA will keep eliciting the missing value until it invokes whatever provisions the platform and application have in place for repeated nomatch failures.
• Finally, information might be present in both the VoiceXML and the ASR result, but in inconsistent formats. For example, an ASR grammar might provide a structured object for the drink which includes the size and whether the drink is diet or non-diet, but the VoiceXML form might only expect a string consisting of the name of the drink. The system's behaviour in these situations would depend on what is being done with the results. For example, a structured object might be sent to a server side script that's expecting a string, and the consequences of this would depend on the server script.

In order to address these potential problems, the committee is looking at various approaches to ensuring consistency between the grammar and the VoiceXML.

4. System Output

4.1 Prompts

The prompt element controls the output of synthesized speech and prerecorded audio. Conceptually, prompts are instantaneously queued for play, so interpretation proceeds until the user needs to provide an input. At this point, the prompts are played, and the system waits for user input. Once the input is received from the speech recognition subsystem (or the DTMF recognizer), interpretation proceeds.

Prompts have the following attributes:

4.1.1 Basic Prompts

You’ve seen prompts in the previous examples:

<prompt>Please say your city.</prompt>

You can leave out the <prompt> ... </prompt> if:

• There is no need to specify a prompt attribute (like bargein), and

• The prompt consists entirely of PCDATA (contains no speech markups) or consists of just an <audio> or <value> element.

For instance, these are also prompts:

Please say your city.

<audio src="say_your_city.wav"/>

But in this example, the enclosing prompt elements are required due to the embedded speech markups:

<prompt>Please <emphasis>say</emphasis> your city.</prompt>

4.1.2 Speech Markup

VoiceXML uses the speech markup elements defined in the W3C Speech Synthesis Markup Language.

Prompts can include special controls to affect how a speech synthesis engine audibly renders text. Such controls are pronunciation, rate, pitch, and other characteristics. For example, emphasis and timing can be controlled as:

<prompt> This is <emphasis>computer-generated</emphasis> text.
   <break size="medium"/> Do you like it? </prompt>

These controls, as well as language information, are inherited down the document hierarchy.

The VoiceXML platform should be a Conforming Speech Synthesis Markup Language Processor as defined in the W3C Speech Synthesis Markup Language. While this requires a platform to process documents with one or more "xml:lang" attributes defined, it does not require that the platform must be multi-lingual. When an unsupported language is encountered, the platform throws an error.unsupported.language event which specifies the language in its message variable.

Isssues:

• It is intended that synthesis elements described in this document are synchronized with the Speech Synthesis Markup Language. Where there are discrepanies during the W3C process, definitions in the Speech Synthesis Markup Language take precedence.
• A future version of this document may use a schema approach which specifies synthesis elements by reference to a separate namespace defined in the Speech Synthesis Markup Language.

4.1.2.1 <break>

Specifies a pause in the speech output. Attributes of <break> are:

At most one of time and size must be specified. If neither are specified, size="medium" is assumed.

4.1.2.2 <paragraph>

Identifies the enclosed text as a paragraph, containing zero or more sentences. For brevity, <p> is supported as an equivalent of <paragraph>.

Attributes of <paragraph> and <p> are:

4.1.2.3 <sentence>

Identifies the enclosed text as a sentence. For brevity, <s> is supported as an equivalent of <sentence>.

Attributes of <sentence> and <s> are:

4.1.2.4 <emphasis>

Specifies that the enclosed text should be spoken with emphasis. Attributes of <emphasis> are:

4.1.2.5 <prosody>

Specifies prosodic information for the enclosed text.

Attributes of prosody are:

4.1.2.6 <say-as>

Specifies the type of text construct contained within the element. Attributes of <say-as> are:

The values of "type" are a text type optionally followed by a colon and format:

For example:

<say-as type="date:ymd"> 2000/1/20 </say-as>

would be spoken as "January 20th two thousand".

4.1.2.7 <phoneme>

Specifies a phonetic pronunciation for the contained text. Attributes for <phoneme> are:

4.1.2.8 <voice>

Specifies voice characteristics for the spoken text. Attributes for <voice> are:

4.1.2.9 <mark>

This feature should be ignored by VoiceXML platforms.

Issues:

• The utility of this element, as well as how notifications could be reported in VoiceXML, is being investigated.

4.1.3 Audio Prompting

Prompts can consist of any combination of prerecorded files, audio streams, or synthesized speech:

<prompt>
   Welcome to the Bird Seed Emporium.
   <audio src="rtsp://www.birdsounds.example.com/thrush.wav"/>
   We have 250 kilogram drums of thistle seed for
   <say-as class="currency">$299.95</say-as>
   plus shipping and handling this month.
   <audio src="http://www.birdsounds.example.com/mourningdove.wav"/>
</prompt>

Audio can be played in any prompt. Typically it is specified via a URI, but it can also be in an audio variable previously recorded:

<prompt>
   Your recorded greeting is
   <value expr="greeting"/>
   To rerecord, press 1.
   To keep it, press pound.
   To return to the main menu press star M.
   To exit press star, star X.
</prompt>

The audio element can have alternate content in case the audio sample is not available:

<prompt>
   <audio src="welcome.wav"> 
     <emphasis>Welcome</emphasis> to the Voice Portal. 
   </audio>
</prompt>

If the audio file cannot be played (e.g. unsupported format, invalid URI, etc.), the content of the audio element is played instead. The content may include text, speech markup, or another audio element. If the audio file cannot be played and the content of the audio element is empty, an appropriate error event will be thrown.

If <audio> contains an 'expr' attribute evaluating to null, then the element, including its alternate content, is ignored. This allows a developer to specify <audio> elements with dynamically assigned content which, if the element is not required, can be ignored by assigning its 'expr' a null value. For example, the following code shows how this could be used to play back a hand of cards using concatenated audio clips:

<form>
  <!-- script contains the function sayCard(position,type) which returns
       audio or textual description of the card in the specified position; 
       if there are no more cards, then returns null --!>
  <script src="cardgame.js"/>

  <field name="takecard" type="boolean">
       <prompt>
           <audio src="you_have.wav">You have the following cards: </audio>
           <!-- maximum of hand of 5 cards is described --!>
           <audio expr="sayCard(audio,1)"><value expr="sayCard(text,1)"/></audio>
           <audio expr="sayCard(audio,2)"><value expr="sayCard(text,2)"/></audio>
           <audio expr="sayCard(audio,3)"><value expr="sayCard(text,3)"/></audio>
           <audio expr="sayCard(audio,4)"><value expr="sayCard(text,4)"/></audio>
           <audio expr="sayCard(audio,5)"><value expr="sayCard(text,5)"/></audio>
           <audio src="another.wav">Would you like another card?</audio>
        </prompt>
         <filled>
              <if cond="takecard">
                  <script>takeAnotherCard()</script>
                  <clear/>
              <else/>
                  <goto next="./make_bid.html"/>
              </if>
         </filled>
   </field>
</form>

Attributes of <audio> include:

Issues:

• A future revision of this document may add a boolean "stream" attribute to <audio> indicating that the interpreter is to begin processing the audio content as it arrives and not to wait for full retrieval. This will give developers explicit control over audio streaming. Until then, platforms may choose to stream audio as an optimization.

4.1.4 <value> Element

The <value> element is used to insert the value of an expression into a prompt. For example if n is 12, the prompt

<prompt>
  <value expr="n*n"/> is the square of <value expr="n"/>.
</prompt>

will result in the text string "144 is the square of 12" being passed to the speech synthesis engine.

When the 'expr' attribute in <value> specifies a <record> field item variable, the recorded audio is played back (see example in Section 2.3.6).

The manner in which the value attribute is played is controlled by the surrounding speech synthesis markup. For instance, a value can be played as a date in the following example:

<var name="date" expr="'2000/1/20'"/>
<prompt>
    <say-as type="date:ymd"> <value expr="date"> </say-as>
</prompt>

Evaluation of the 'expr' attrbute may return a string containing speech synthesis markup (as defined by SSML). Thus it may return any of the structures that can be contained within a prompt element except another value element (which is not defined within SSML).

The platform must perform the following steps to convert a prompt containing values returning one or more SSML expressions.

1. Expand each value expression by replacing the value element by the text produced by the expression. (Do not recursively expand value elements.)
2. Ensure that expanded prompt is a valid prompt structure. If not, throw an error.semantic event.

Consider the following assignments to three ECMAScript variables

    e1 = "<voice gender='male' age='20'>";
    e2 = "<paragraph>I want a <emphasis>large</emphasis> pizza with pepperoni</paragraph>"
    e3 = "</voice>";

and the following prompt element containing a sequence of value elements.

<prompt>
    <value expr="e1"/>
    <value expr="e2"/>
    <value expr="e3"/>
</prompt>

After expansion and validation this prompt is equivalent to the following prompt element. Note that e1 and e3 containing matching opening and closing voice elements so that after expansion the prompt remains well-formed XML.

<prompt>
    <voice gender='male' age='20'>
    <paragraph>I want a <emphasis>large</emphasis> pizza with pepperoni</paragraph>
    </voice>
</prompt>

Issues:

• Issue 1: XML Special characters in prompt CDATA
  Consider the following ECMAScript variable assignment

    e1 = '<emphasis>$12345</emphasis>'
  

  And the following VoiceXML fragment

    <prompt>The price of AT&amp;T is <value expr="e1"/>.</prompt>
  

  The initial XML parse of the containing VoiceXML document turns AT&amp;T into AT&T. However, since '&' is a reserved character in XML, it must be re-quoted by the VoiceXML interpreter before re-parsing. This must result in the following text being sent to the SSML interpreter:

    The price of AT&amp;T is <emphasis>$12345</emphasis>.
  

  Implication: any XML special characters contained in CDATA within the subtree of a prompt element must be handled specially by the VoiceXML interpreter.
   
• Issue 2: XML Special characters returned in string of value expression
  Since the string returned by the expression of a value element is re-parsed as XML, any XML special characters contained within the string must be quoted by the VoiceXML application developer. For example, when the following variable assignment

    e1 = 'AT&T';
  

  is referenced in a prompt element as

    <prompt>The price of <value expr="e1"/> is $12345.</prompt>
  

  this would result in the following text being sent to the SSML parser. Note that the entity references are properly formatted making it a well-formed XML fragment.

   The price of AT&T is $1.
  

  Negative example: The following is an example of an expected mis-handling on entity references. This following is legal ECMAScript however once the string is substituted into the prompt the XML is no longer well-formed.

    e1 = 'AT&T';
  

  This is the same prompt element.

    <prompt>The price of <value expr="e1"/> is $12345.</prompt>
  

  Resulting in the followong ill-formed XML being sent to the SSML parser:

    The price of AT&T is $1.
  

  We encourage reviewer comment on the trade-off between the expressive power of a value expression returning SSML vs. the potential for developer confusion.

Attributes of <value> are:

4.1.5 Barge-in

If an implementation platform supports barge-in, the service author can specify whether a user can interrupt, or "barge-in" on, a prompt using speech or DTMF input. This speeds up conversations, but is not always desired. If the application author requires that the user must hear all of a warning, legal notice, or advertisement, barge-in should be disabled. This is done with the bargein attribute:

<prompt bargein="false"><audio src="legalese.wav"/></prompt>

Users can interrupt a prompt whose bargein attribute is true, but must wait for completion of a prompt whose bargein attribute is false. In the case where several prompts are queued, the bargein attribute of each prompt is honored during the period of time in which that prompt is playing. If bargein occurs during any prompt in a sequence, all subsequent prompts are not played. If the bargein attribute is not specified, then the value of the bargein property is used if set.

When the bargein attribute is false, any DTMF input buffered in a transition state is deleted from the buffer (Section 4.1.8 describes input collection during transition states).

Note that not all speech recognition engines or implementation platforms support barge-in. For a platform to support barge-in, it must support at least one of the barge-in types described in Section 4.1.5.1.

4.1.5.1 Barge-in type

When barge-in is enabled, the bargeintype attribute can be used to suggest the type of bargein the platform will perform. Possible values for this attribute are:

If the bargeintype attribute is not specified, then the value of the bargeintype property is used if set.

Implementations that claim to support barge-in are required to support at least one of these three types. For each unsupported type x, a platform is required to document which of the other two types will be used when type x is requested.

Note that with energy and speech barge-in, the prompt is stopped irrespective of whether the recognizer matches the utterance/DTMF string or not. For the recognition type of barge-in, a 'nomatch' event will never be generated.

Mixing these types within a single queue of prompts can result in unpredictable behavior and is discouraged. Where possible, platforms should attempt to behave in a reasonable manner.

4.1.6 Prompt Selection

Tapered prompts are those that may change with each attempt. Information-requesting prompts may become more terse under the assumption that the user is becoming more familiar with the task. Help messages become more detailed perhaps, under the assumption that the user needs more help. Or, prompts can change just to make the interaction more interesting.

Each form item and each menu has an internal prompt counter that is reset to one each time the form or menu is entered. Whenever the system uses a prompt, its associated prompt counter is incremented. This is the mechanism supporting tapered prompts.

For instance, here is a form with a form level prompt and field level prompts:

<form id="tapered">
  <block>
    <prompt bargein="false">
      Welcome to the ice cream survey.
    </prompt>
  </block>
  <field name="flavor">
  <grammar mode="voice">
    <rule id="root" scope="public">
      <one-of>
        <item>vanilla </item>
        <item>chocolate</item>
        <item>strawberry</item>
     </one-of>
    </rule>
   </grammar>
   <prompt count="1">What is your favorite flavor?</prompt>
   <prompt count="3">Say chocolate, vanilla, or strawberry.</prompt>
   <help>Sorry, no help is available.</help>
  </field>
</form>

A conversation using this form follows:

C: Welcome to the ice cream survey.

C: What is your favorite flavor? (the "flavor" field’s prompt counter is 1)

H: Pecan praline.

C: I do not understand.

C: What is your favorite flavor? (the prompt counter is now 2)

H: Pecan praline.

C: I do not understand.

C: Say chocolate, vanilla, or strawberry. (prompt counter is 3)

H: What if I hate those?

C: I do not understand.

C: Say chocolate, vanilla, or strawberry. (prompt counter is 4)

H: ...

When it is time to select a prompt, the prompt counter is examined. The child prompt with the highest count attribute less than or equal to the prompt counter is used. If a prompt has no count attribute, a count of "1" is assumed.

A conditional prompt is one that is spoken only if its condition is satisfied. In this example, a prompt is varied on each visit to the enclosing form.

<form id="another_joke">
   <var name="r" expr="Math.random()"/>
   <field name="another" type="boolean">
       <prompt cond="r < .50">
          Would you like to hear another elephant joke?
       </prompt>
       <prompt cond="r >= .50">
         For another joke say yes.  To exit say no.
       </prompt>
       <filled>
          <if cond="another">
            <goto next="#pick_joke"/>
          </if>
       </filled>
   </field>
</form>

When a prompt must be chosen, a set of prompts to be queued is chosen according to the following algorithm:

1. Form an ordered list of prompts consisting of all prompts in the enclosing element in document order.
2. Remove from this list all prompts whose cond evaluates to false.
3. Find the "correct count": the highest count among the prompt elements still on the list less than or equal to the current count value.
4. Remove from the list all the elements that don't have the "correct count".

All elements that remain on the list will be queued for play.

4.1.7 Timeout

The timeout attribute specifies the interval of silence allowed while waiting for user input after the end of the last prompt. If this interval is exceeded, the platform will throw a noinput event. This attribute defaults to the value specified by the timeout property (see Section 6.3.4).

The reason for allowing timeouts to be specified as prompt attributes is to support tapered timeouts. For example, the user may be given five seconds for the first input attempt, and ten seconds on the next.

The prompt timeout attribute determines the noinput timeout for the following input:

<prompt count="1">
  Pick a color for your new Model T.
</prompt>

<prompt count="2" timeout="120s">
  Please choose color of your new nineteen twenty four
  Ford Model T. Possible colors are black, black, or
  black.  Please take your time.
</prompt>

If several prompts are queued before a field input, the timeout of the last prompt is used.

4.1.8 Prompt Queueing and Input Collection

A VoiceXML interpreter is at all times in one of two states:

• waiting for input in a field item (such as <field>, <record>, or <transfer>), or
• transitioning between field items in response to an input (including spoken utterances, dtmf key presses, and input-related events such as a noinput or nomatch event) received while in the waiting state. While in the transitioning state no speech input is collected, accepted or interpreted. Consequently root and document level speech grammars (such as defined in <link>s) may not be active at all times. However, DTMF input (including timing information) should be collected and buffered in the transition state.

The waiting and transitioning states are related to the phases of the Form Interpretation Algorithm as follows:

• the waiting state is entered in the collect phase of a field item, and
• the transitioning state encompasses the process and select phases, and the collect phase for non-field items (such as <block>s).

This distinction of states is made in order to greatly simplify the programming model. In particular, an important consequence of this model is that the VoiceXML application designer can rely on all executable content (such as the content of <filled> and <block> elements) being run to completion, because it is executed while in the transitioning state, which may not be interrupted by input.

While in the transitioning state various prompts are queued, either by the <prompt> element in executable content or by the <prompt> element in field items. In addition, audio may be queued by the fetchaudio attribute. The queued prompts and audio are played either

• when the interpreter reaches the waiting state, at which point the prompts are played and the interpreter listens for input that matches one of the active grammars, or
• when the interpreter begins fetching a resource (such as a document) for which fetchaudio was specified. In this case the prompts queued before the fetchaudio are played to completion, and then, if the resouce actually needs to be fetched (i.e. it is not unexpired in the cache), the fetchaudio is played until the fetch completes. The interpreter remains in the transitioning state and no input is accepted during the fetch.

Before the interpreter exits all queued prompts are played to completion. The interpreter remains in the transitioning state and no input is accepted while the interpreter is exiting.

It is a permissible optimization to begin playing prompts queued during the transitioning state before reaching the waiting state, provided that correct semantics are maintained regarding processing of the input audio received while the prompts are playing, for example with respect to bargein and grammar processing.

The following examples illustrate the operation of these rules in some common cases.

Case 1

Typical non-fetching case: field, followed by executable content (such as <block> and <filled>), followed by another field.

in document d0

    <field name="f0"/>

    <block>
        executable content e1
        queues prompts {p1}
    </block>

    <field name="f2">
        queues prompts {p2}
        enables grammars {g2}
    </field>

As a result of input received while waiting in field f0 the following actions take place:

• in transitioning state
  • execute e1 (without goto)
  • queue prompts {p1}
  • queue prompts {p2}
• in waiting state, simultaneously
  • play prompts {p1,p2}
  • enable grammars {g2}
  • wait for input

Case 2

Typical fetching case: field, followed by executable content (such as <block> and <filled>) ending with a <goto> that specifies fetchaudio, ending up in a field in a different document that is fetched from a server.

in document d0

    <field name="f0"/>

    <block>
        executable content e1
        queues prompts {p1}
        ends with goto f2 in d1 with fetchaudio fa
    </block>

in document d1

    <field name="f2">
        queues prompts {p2}
        enables grammars {g2}
    </field>

As a result of input received while waiting in field f0 the following actions take place:

• in transitioning state
  • execute e1
  • queue prompts {p1}
  • simultaneously
    • fetch d1
    • play prompts {p1} to completion and then play fa until fetch completes
  • queue prompts {p2}
• in waiting state, simultaneously
  • play prompts {p2}
  • enable grammars {g2}
  • wait for input

Case 3

As in Case 2, but no fetchaudio is specified.

in document d0

    <field name="f0"/>

    <block>
        executable content e1
        queues prompts {p1}
        ends with goto f2 in d1 (no fetchaudio specified)
    </block>

in document d1

    <field name="f2">
        queues prompts {p2}
        enables grammars {g2}
    </field>

As a result of input received while waiting in field f0 the following actions take place:

• in transitioning state
  • execute e1
  • queue prompts {p1}
  • fetch d1
  • queue prompts {p2}
• in waiting state, simultaneously
  • play prompts {p1, p2}
  • enable grammars {g2}
  • wait for input

5. Control flow and scripting

5.1 Variables and Expressions

VoiceXML variables are in all respects equivalent to ECMAScript variables: they are part of the same variable space. VoiceXML variables can be used in a <script> just as variables defined in a <script> can be used in VoiceXML. Declaring a variable using <var> is equivalent to using a 'var' statement in a <script> element. <script> can also appear everywhere that <var> can appear.

The variable naming convention is as in ECMAScript, but names beginning with the underscore character ("_") and names ending with a dollar sign ("$") are reserved for internal use. VoiceXML variables, including field item variables, must not contain ECMAScript reserve words. They must also follow ECMAScript rules for referential correctness. For example, variable names must be unique and their declaration must not include a dot - "var x.y" is an illegal declaration in ECMAScript. Variable names which violate ECMAScript rules cause an 'error.semantic' event to be thrown.

5.1.1 Declaring Variables

Variables are declared by <var> elements:

<var name="home_phone"/> 
<var name="pi" expr="3.14159"/> 
<var name="city" expr="'Sacramento'"/>

They are also declared by form items:

<field name="num_tickets" type="number"> 
   <prompt>How many tickets do you wish to purchase?</prompt> 
</field>

Variables declared without an explicit initial value are initialized to the ECMAScript undefined value. Variables must be declared before being used either in VoiceXML or ECMAScript. Variables declared using "var" in ECMAScript can be used in VoiceXML, just as declared VoiceXML variables can be used in ECMAScript. ECMAScript allows the use of undeclared variables; however, because these variables are global, they must not be used from VoiceXML. (VoiceXML implementations are free to restrict the use of undeclared variables, even from within ECMAScript blocks like <script>.). Explicit <var> declarations in VoiceXML aid readibility and application flow analysis.

In a form, the variables declared by <var> and those declared by form items are initialized when the form is entered. The initializations are guaranteed to take place in document order, so that this, for example, is legal:

<form id="test"> 
   <var name="one" expr="1"/> 
   <field name="two" expr="one+1" type="number"></field> 
   <var name="three" expr="two+1"/> 
   <field name="go_on" type="boolean"> 
     <prompt>Say yes or no to continue</prompt> 
   </field> 
   <filled> 
       <goto next="#tally"/> 
   </filled> 
</form>

When the user visits this <form>, the form’s initialization first declares the variable one and sets its value to 1. Then it declares the field item variable two and gives it the value 2. Then the initialization logic declares the variable three and gives it the value 3. The form interpretation algorithm then enters its main interpretation loop and begins at the go_on field.

5.1.2 Variable Scopes

Variables can be declared in following scopes:

The following diagram shows the scope hierarchy:

Figure 7: The scope hierarchy.

The curved arrows in this diagram show that each scope contains a pre-defined variable whose name is the same as the scope that refers to the scope itself. This allows you for example in the anonymous, dialog, and document scopes to refer to a variable X in the document scope using document.X. As another example, a <filled>'s variable scope is an anonymous scope local to the <filled>, whose parent variable scope is that of the <form>.

It is not recommended to use "session", "application", "document", and "dialog" as the names of variables and form items. While they are not reserved words, using them hides the pre-defined variables with the same name because of ECMAScript scoping rules used by VoiceXML.

5.1.3 Referencing Variables

Variables are referenced in cond and expr attributes:

<if cond="city == 'LA'"> 
   <assign name="city" expr="'Los Angeles'"/> 
<elseif cond="city == 'Philly'"/> 
   <assign name="city" expr="'Philadelphia'"/> 
<elseif cond="city =='Constantinople'"/> 
   <assign name="city" expr="'Istanbul'"/> 
</if> 

<assign name="var1" expr="var1 + 1"/> 

<if cond="i > 1"> 
   <assign name="i" expr="i-1"/> 
</if>

The expression language used in cond and expr is precisely ECMAScript. Note that the cond operators "<", "<=", and "&&" must be escaped in XML (to "&lt;" and so on). For clarity, examples in this document do not use XML escapes.

Variable references match the closest enclosing scope according to the scope chain given above. You can prefix a reference with a scope name for clarity or to resolve ambiguity. For instance to save the value of a form field item variable for use later on in a document:

<assign name="document.ssn" expr="dialog.ssn"/>

If the application root document has a variable x, it is referred to as application.x in non-root documents, and either application.x or document.x in the application root document.

5.1.4 Standard Session Variables

session.telephone.ani

Automatic Number Identification. This variable provides the result from the Automatic Number Identification service that provides the receiver of a telephone call with the number of the calling phone. This information is provided only if the service is supported, and is undefined otherwise.

session.telephone.dnis

Dialed Number Identification Service. This variable provides the result from the Dialed Number Identification Service that identifies for the receiver of a call the number that the caller dialed. This information is provided only if the service is supported, and is undefined otherwise.

session.telephone.iidigits

Information Indicator Digits. This variable provides information about the originating line (e.g. payphone, cellular service, special operator handling, prison) of the caller. Telecordia publishes the complete list of II digits in Section 1 of each volume of the "Local Exchange Routing Guide". This information is provided only if the service is supported, and is undefined otherwise.

session.telephone.uui

User to User Information. This variable returns supplementary information provided as part of an ISDN call set-up from a calling party. This information is provided only if the service is supported, and is undefined otherwise.

session.telephone.rdnis

Redirect Dialed Number Information Service. This variable provides the number from which a call diversion or transfer was invoked. This variable is undefined if the number is unavailable. For example, suppose person A subscribes to a voice messaging service, and forwards all calls to a voice server. Person B then calls A and gets routed to the voice server. A VoiceXML application on the voice server should see RDNIS as A's number, while DNIS is the number of the voice server, and ANI is B's number.

session.telephone.redirect_reason

Redirect Reason. This variable is a string that indicates the reason the call was redirected to this server. This variable is always available if session.telephone.rdnis is available. It is undefined if session.telephone.rdnis is undefined. The following reason codes are defined: "unknown", "user-busy", "no-answer", "unavailable", "unconditional", "time-of-day", "do-not-disturb", "deflection", "follow-me", "out-of-service", "away".

Issues:

• The session.telephone variable space may be deprecated in a future version of this document in favor of the connection variable space defined below. The connection variable space has the advantages that it provides an extensible attribute space for protocol-specific information while providing a generic core of information that will be available over all protocols. The core object attributes are designed to be independent of whether the call was received or placed.

  session.connection.local.uri

  This variable is a URI which addresses the local interpreter context device.

  session.connection.remote.uri

  This variable is a URI which addresses the remote caller device. (For instance, a telephone device should be addressed with a tel: URL syntax conforming to RFC 2806.)

  session.connection.protocol.name

  This variable is the name of the connection protocol. The name also represents the subobject name for protocol specific information. For instance, if session.connection.protocol.name is 'q931', session.connection.protocol.q931.uui might specify the user-to-user information property of the connection.

  session.connection.protocol.version

  This variable is the version of the connection protocol.

  Event names related to telephony (e.g. telephone.disconnect.hangup, error.telephone.noauthorization) may also be changed to be more agnostic to connection type.

5.1.5 Standard Application Variables

application.lastresult$

This read-only variable holds information about the last recognition to occur within this application. It is an array of elements, where each element application.lastresult$[i] represents a possible result through the following read-only variables:

application.lastresult$[i].confidence

The confidence level for this interpretation from 0.0-1.0. A value of 0.0 indicates minimum confidence, and a value of 1.0 indicates maximum confidence. More specific interpretation of a confidence value is platform-dependent.

application.lastresult$[i].utterance

The raw string of words that were recognized for this interpretation. The exact tokenization and spelling is platform-specific (e.g. "five hundred thirty" or "5 hundred 30" or even "530").

application.lastresult$[i].inputmode

For this interpretation,the mode in which user input was provided: dtmf or voice.

application.lastresult$[i].interpretation

An ECMAscript variable containing the interpretation as described in Section 3.1.5.

Interpretations are sorted by confidence score, from highest to lowest. Different elements in application.lastresult$ will always differ in their utterance, interpretation, or both.

The number of application.lastresult$ elements is guaranteed to be greater than or equal to one and less than or equal to the system property "maxnbest". If no results have been generated by the system, then "application.lastresult$" shall be ECMAScript undefined.

Additionally, application.lastresult$ itself contains the properties confidence, utterance, inputmode, and interpretation corresponding to those of the 0th element in the array.

All of the shadow variables described above are set immediately after any recognition. When an application root document is loaded, this variable is set to the value undefined.

The following example show how application.lastresult$ can be used in a field level <catch> to access a <link> grammar recognition result and transition to different dialog states depending on confidence:

<link event="menulinkevent">
    <grammar src="./linkgrammar.grxml" type="application/grammar+xml"/>
</link>

<form>
    ...
    <field>
        ...
        <catch event="menulinkevent">
            <if cond="application.lastresult$.confidence < 0.7">
        <goto nextitem="confirmlinkdialog"/>
        <else/>
        <goto next="./main_menu.html"/>
        </if>
        </catch>
    </field>
    ...
</form>

5.2 Event Handling

The platform throws events when the user does not respond, doesn't respond intelligibly, requests help, etc. The interpreter throws events if it finds a semantic error in a VoiceXML document, or when it encounters a <throw> element. Events are identified by character strings.

Each element in which an event can occur has a set of catch elements, which include:

• <catch>

• <error>

• <help>

• <noinput>

• <nomatch>

An element inherits the catch elements ("as if by copy") from each of its ancestor elements, as needed. If a field, for example, does not contain a catch element for nomatch, but its form does, the form’s nomatch catch element is used. In this way, common event handling behavior can be specified at any level, and it applies to all descendents.

The "as if by copy" semantics for inheriting catch elements implies that when a catch element is executed, variables are resolved and thrown events are handled relative to the scope where the original event originated, not relative to the scope that contains the catch element. For example, consider a catch element that is defined at document scope handling an event that originated in a <field> within the document. In such a catch element variable references are resolved relative to the <field>'s scope, and if an event is thrown by the catch element it is handled relative to the <field>. Similarly, relative URL references in a catch element are resolved against the active document and not relative to the document in which they were declared.

5.2.1 Throw

The <throw> element throws an event. These can be the pre-defined ones:

<throw event="nomatch"/> 
<throw event="telephone.disconnect.hangup"/>

or application-defined events:

<throw event="com.att.portal.machine"/>

Attributes of <throw> are:

Unless explicited stated otherwise, VoiceXML does not specify when events are thrown.

5.2.2 Catch

The catch element associates a catch with a document, dialog, or form item. It contains executable content.

<form id="launch_missiles"> 
  <field name="password"> 
    <prompt>What is the code word?</prompt> 
    <grammar>
     <rule id="root" scope="public">rutabaga</rule>
    </grammar>
   <help>It is the name of an obscure vegetable.</help> 
   <catch event="nomatch noinput" count="3"> 
     <prompt>Security violation!</prompt> 
     <submit next="apprehend_felon" namelist="user_id"/> 
   </catch> 
  </field> 
  <block> 
   <goto next="#get_city"/> 
  </block> 
</form>

The catch element's anonymous variable scope includes the special variable _event which contains the name of the event that was thrown. For example, the following catch element can handle two types of events:

<catch event="event.foo event.bar">
  <if cond="_event=='event.foo'">
    <!-- Play this for event.foo events -->
    <audio src="foo.wav"/>
  <else/>
    <!-- Play this for event.bar events -->
    <audio src="bar.wav"/>
  </if>
  <!-- Continue with common handling for either event -->
</catch>

The _event variable is inspected to select the audio to play based on the event that was thrown. The foo.wav file will be played for event.foo events. The bar.wav file will be played for event.bar events. The remainder of the catch element contains executable content that is common to the handling of both event types.

The catch element's anonymous variable scope also includes the special variable _message which contains the value of the message string from the corresponding <throw> element, or a platform-dependent value for the pre-defined events raised by the platform. If the thrown event does not specify a message, the value of _message is ECMAScript undefined.

There are no inherent limitations on the executable context in a <catch> element: execution proceeds as normal until the interpreter exits. Consider the following example where the application cleans up after a user hangup by submitting information to a server:

<catch event="telephone.disconnect.hangup">
    <submit namelist="myExit" next="http://mysite/exit.jsp"/>
</catch>

The document returned by the server is processed normally by the interpreter until it reaches the first point in the document when input or output is required. At that point, the interpreter exits.

If a <catch> element contains a <throw> element with the same event, then there may be an infinite loop:

<catch event="help"> 
   <throw event="help"/> 
</catch>

A platform could detect this situation and throw a semantic error instead.

Attributes of <catch> are:

5.2.3 Shorthand Notation

The <error>, <help>, <noinput>, and <nomatch> elements are shorthands for very common types of <catch> elements.

The <error> element is short for <catch event="error"> and catches all events of type error:

<error>
  An error has occurred -- please call again later.
  <exit/>
</error>

The <help> element is an abbreviation for <catch event="help">:

<help>No help is available.</help>

The <noinput> element abbreviates <catch event="noinput">:

<noinput>I didn't hear anything, please try again.</noinput>

And the <nomatch> element is short for <catch event="nomatch">:

<nomatch>I heard something, but it wasn't a known city.</nomatch>

These elements take the attributes:

5.2.4 Catch Element Selection

An element inherits the catch elements ("as if by copy") from each of its ancestor elements, as needed. For example, if a <field> element inherits a <catch> element from the document

<catch event="event.foo">
    <audio src="beep.wav"/>
</catch>


<form>
    <field>
        ...
        <nomatch>
            <throw event="event.foo"/>
        </nomatch>
    </field>
</form>

then the <catch> element is implicitly copied into <field> as if defined below:

<form>
    <field>
        ...
        <nomatch>
            <throw event="event.foo"/>
        </nomatch>
        <catch event="event.foo">
            <audio src="beep.wav"/>
        </catch>
    </field>
</form>

When an event is thrown, the scope in which the event is handled and its enclosing scopes are examined to find the best qualified catch element, according to the following algorithm:

1. Form an ordered list of catches consisting of all catches in the current scope and all enclosing scopes (form item, form, document, application root document, interpreter context), ordered first by scope (starting with the current scope), and then within each scope by document order.
2. Remove from this list all catches whose event name does not match the event being thrown or whose cond evaluates to false.
3. Find the "correct count": the highest count among the catch elements still on the list less than or equal to the current count value.
4. Select the first element in the list with the "correct count".

The name of a thrown event matches the catch element event name if it is either an exact match or a prefix match. A prefix match occurs when the catch element event attribute is a token prefix of the name of the event being thrown, where the dot is the token separator, all trailing dots are removed, and the empty string matches everything. For example,

<catch event="telephone.disconnect">

will prefix match the event telephone.disconnect.transfer.

<catch event="com.example.myevent">

prefix matches com.example.myevent.event1., com.example.myevent. and com.example.myevent..event1 but not com.example.myevents.event1. Finally,

<catch event="">

prefix matches all events.

Note that the catch element selection algorithm gives priority to catch elements that occur earlier in a document over those that occur later, but does not give priority to catch elements that are more specific over those that are less specific. Therefore is generally advisable to specify catch elements in order from more specific to less specific. For example, it would be advisable to specify catch elements for "error.foo" and "error" in that order, as follows:

<catch event="error.foo"> 
  ...
</catch> 
<catch event="error"> 
  ... 
</catch>

If the catch elements were specified in the opposite order, the catch element for "error.foo" would never be executed.

5.2.5 Default Catch Elements

The interpreter is expected to provide implicit default catch handlers for the noinput, help, nomatch, cancel, exit, and error events if the author did not specify them.

The system default behavior of catch handlers for various events and errors is summarized by the definitions below that specify (1) whether any audio response is to be provided, and (2) how execution is affected. Note: where an audio response is provided, the actual content is platform dependent.

Specific platforms and locales will differ in the default prompts presented.

5.2.6 Event Types

There are pre-defined events and application-defined events. Events are also subdivided into plain events (things that happen normally), and error events (abnormal occurrences). The error naming convention allows for multiple levels of granularity.

The pre-defined events are:

cancel

The user has requested to cancel playing of the current prompt.

telephone.disconnect.hangup

The user has hung up.

telephone.disconnect.transfer

The user has been transferred unconditionally to another line and will not return.

exit

The user has asked to exit.

help

The user has asked for help.

noinput

The user has not responded within the timeout interval.

nomatch

The user input something, but it was not recognized.

maxspeechtimeout

The user input was too long exceeding the 'maxspeechtimeout' property.

The predefined errors are:

error.badfetch

The interpreter context throws this event when a fetch of a document has failed and the interpreter context has reached a place in the document interpretation where the fetch result is required.  Fetch failures result from missing documents, malformed URIs, communications errors, timeouts, security violations, or malformed documents.

If the interpreter context has speculatively prefetched a document and that document turns out not to be needed, error.badfetch is not thrown.  Likewise if the fetch of an <audio> document fails and if there is a nested alternate <audio> document whose fetch then succeeds, or if there is nested alternate text, no error.badfetch occurs.

When an interpreter context is transitioning to a new document, and that document has been fetched but has syntax or semantic errors, the interpreter context throws error.badfetch, but again only at the point in time where the new document is actually needed, not before.

error.badfetch.http.<response code>
error.badfetch.protocol.<response code>

In the case of a fetch failure, the interpreter context must use a detailed event type telling which specific HTTP or other protocol-specific response code was encountered. The value of the response code for HTTP is defined in HTTP 1.1. This allows applications to differentially treat a missing document from a prohibited document, for instance. The value of the response code for other protocols (such as HTTPS, RTSP, and so on) is dependent upon the protocol.

error.semantic

A run-time error was found in the VoiceXML document, e.g. a divide by 0, substring bounds error, or an undefined variable was referenced.

error.noauthorization

The user is not authorized to perform the operation requested (such as dialing an invalid telephone number, or one which the user is not allowed to call).

error.unsupported.format

The requested resource has a format that is not supported by the platform, e.g. an unsupported grammar format, audio file format, object type, or media type.

error.unsupported.language

The platform does not support the language for either speech synthesis or speech recognition.

error.unsupported.element

The platform does not support the given element. For instance, if a platform does not implement <transfer>, it must throw error.unsupported.transfer. This allows an author to use event handling to adapt to different platform capabilities.

Errors encountered during document loading, including transport errors (no document found, HTTP status code 404, and so on) and syntactic errors (no XML header, no <vxml> element, etc) result in a badfetch error event raised in the calling document, while errors after loading, such as semantic errors, are raised in the document itself.

Application-specific error types should follow the following format:

error.com.mot.mix.noauth

Access to personal profile information is not authorized.

error.com.ibm.portal.restricted

The document tried to access a restricted resource.

Catches can catch specific events (cancel) or all those sharing a prefix (error.unsupported).

Issues:

• The determination of a badfetch for all fetchable items (grammars, prompts, VoiceXML documents, and objects) needs to be clarified. For example, what is the behavior for grammar documents? At which point is a grammar document considered "fetched"? If it has a syntax error, is that a badfetch?
• if fetching takes place late, where is the error handled?
• We are investigating methods of providing more information about the badfetch in cases such as parse failures or inappropriate retrievals (e.g. grammar document returned instead of VoiceXML document). One method would be to encode further information in the error name; for example "error.badfetch.grammar.syntax". A limitation is that multiple errors cannot be reported at the same time. An alternative method we are investigating is to fill a session-level ECMAScript variable with detailed error information.

5.3 Executable Content

Executable content refers to a block of procedural logic. Such logic appears in:

• The <block> form item.

• The <filled> actions in forms and form items.

• Event handlers (<catch>, <help>, et cetera).

This section covers the elements that can occur in executable content.

5.3.1 VAR

This element declares a variable. It can occur in executable content or as a child of <form> or <vxml>. Examples:

<var name="phone" expr="6305551212"/> 
<var name="y" expr="document.z+1"/>

If it occurs in executable content, it declares a variable in the anonymous scope associated with the enclosing <block>, <filled>, or catch element. This declaration is made only when the <var> element is executed. If the variable is already declared in this scope, subsequent declarations act as assignments, as in ECMAScript.

If a <var> is a child of a <form> element, it declares a variable in the dialog scope of the <form>. This declaration is made during the form’s initialization phase as described in Section 6.6.1. The <var> element is not a form item, and so is not visited by the Form Interpretation Algorithm’s main loop.

If a <var> is a child of a <vxml> element, it declares a variable in the document scope. This declaration is made when the document is initialized; initializations happen in document order.

Attributes of <var> include:

5.3.2 ASSIGN

The <assign> element assigns a value to a variable:

<assign name="flavor" expr="'chocolate'"/> 
<assign name="document.mycost" expr="document.mycost+14"/>

Attributes include:

5.3.3 CLEAR

The <clear> element resets one or more form items. Resetting includes:

Setting the form item variable to ECMAScript undefined.

Reinitializing the prompt counter and the event counters for the form item.

For example:

<clear namelist="city state zip"/>

The attribute is:

5.3.4 IF, ELSEIF, and ELSE

The <if> element is used for conditional logic. It has optional <else> and <elseif> elements.

<if cond="total > 1000"> 
  <prompt>This is way too much to spend.</prompt> 
  <throw event="com.xyzcorp.acct.toomuchspent"/> 
</if> 

<if cond="amount < 29.95"> 
  <assign name="x" expr="amount"/> 
<else/> 
  <assign name="x" expr="29.95"/> 
</if> 

<if cond="flavor == 'vanilla'"> 
  <assign name="flavor_code" expr="'v'"/> 
<elseif cond="flavor == 'chocolate'"/> 
  <assign name="flavor_code" expr="'h'"/> 
<elseif cond="flavor == 'strawberry'"/> 
  <assign name="flavor_code" expr="'b'"/> 
<else/> 
  <assign name="flavor_code" expr="'?'"/> 
</if>

5.3.5 PROMPT

Prompts can appear in executable content, in their full generality, except that the <prompt> count attribute is meaningless. In particular, the cond attribute can be used in executable content. Prompts may be wrapped with <prompt> and </prompt>, or represented using PCDATA. Wherever <prompt> is allowed, the PCDATA xyz is interpreted exactly as if it had appeared as <prompt>xyz</prompt>.

<nomatch count="1"> 
  To open the pod bay door, say your code phrase clearly. 
</nomatch> 

<nomatch count="2"> 
  <prompt>
    This is your <emphasis>last</emphasis> chance.
  </prompt> 
</nomatch> 

<nomatch count="3"> 
  Entrance denied. 
  <exit/> 
</nomatch>

5.3.6 REPROMPT

The FIA expects a catch element to queue appropriate prompts in the course of handling an event. Therefore, the FIA does not generally perform the normal selection and queuing of prompts on the next iteration following the execution of a catch element. However, the FIA does perform normal selection and queueing of prompts after the execution of a catch element (<catch>, <error>, <help>, <noinput>, <nomatch>) in two cases:

• if the catch element ends by executing a <goto> to another dialog; in this case the new dialog needs to be guaranteed that its initial prompt remains intact and cannot be suppressed or replaced by a referring dialog; or
• if a <reprompt> is executed in the catch to request that the subsequent prompts be played.

In these two cases, after the FIA selects the next form item to visit, it performs normal prompt processing, including selecting and queuing the form item's prompts and incrementing the form item's prompt counter.

For example, this noinput catch expects the next form item prompt to be selected and played:

<field name="want_ice_cream" type="boolean"> 
  <prompt>Do you want ice cream for dessert?</prompt> 
  <prompt count="2"> 
    If you want ice cream, say yes. 
    If you don’t want ice cream, say no. 
  </prompt> 
  <noinput> 
    I could not hear you. 
    <!-- Cause the next prompt to be selected and played. --> 
    <reprompt/>
  </noinput> 
</field>

A quiet user would hear:

C: Do you want ice cream for dessert?

H: (silence)

C: I could not hear you.

C: If you want ice cream, say yes. If you don’t want ice cream, say no.

H: (silence)

C: I could not hear you.

C: If you want ice cream, say yes. If you don’t want ice cream, say no.

H: No

If there were no <reprompt>, the user would instead hear:

C: Do you want ice cream for dessert?

H: (silence)

C: I could not hear you.

H: (silence)

C: I could not hear you.

H: No

Note that a consequence of skipping the prompt selection phase as described above is that the prompt counter of the form item selected by the FIA after the execution of a catch element (that does not execute a <reprompt> or <goto>) will not be incremented.

Also note that the prompt selection phase following the execution of a catch element (that does not execute a <reprompt> or <goto>) is skipped even if the form item selected by the FIA is different from the previous form item.

5.3.7 GOTO

The <goto> element is used to;

• transition to another form item in the current form,

• transition to another dialog in the current document, or

• transition to another document.

To transition to another form item, use the nextitem attribute, or the expritem attribute if the form item name is computed using an ECMAScript expression:

<goto nextitem="ssn_confirm"/> 
<goto expritem="(type==12)? 'ssn_confirm' : 'reject'"/>

To go to another dialog in the same document, use next (or expr) with only a URI fragment:

<goto next="#another_dialog"/> 
<goto expr="'#' + 'another_dialog'"/>

To transition to another document, use next (or expr) with a URI:

<goto next="http://flight.example.com/reserve_seat"/> 
<goto next="./special_lunch#wants_vegan"/>

The URI may be absolute or relative to the current document. You may specify the starting dialog in the next document using a fragment that corresponds to the value of the id attribute of a dialog. If no fragment is specified, the first dialog in that document is chosen.

Note that transitioning to another dialog in the current document causes the old dialog’s variables to be lost, even in the case where a dialog is transitioning to itself. Transitioning to another document using an absolute or relative URI will likewise drop the old document level variables, even if the new document is the same one that is making the transition. However, document variables are retained when transitioning to an empty URI reference with a fragment identifier. For example, the following statements behave differently in a document with the URI http://someco.example.com/index.vxml:

<goto next="#foo"/>

<goto next="http://someco.example.com/index.vxml#foo"/>

According to RFC 2396, the fragment identifier (the part after the '#') is not part of a URI and transitioning to empty URI references plus fragment identifiers should never result in a new document fetch. Therefore "#foo" in the first statement is an empty URI reference with a fragment identifier and document variables are retained. In the second statement "#foo" is part of a relative URI and the document variables are lost. If you want data to persist across multiple documents, store data in the application scope.

Attributes of <goto> are:

5.3.8 SUBMIT

The <submit> element is similar to <goto> in that it results in a new document being obtained. Unlike <goto>, it lets you submit a list of variables to the document server via an HTTP GET or POST request. For example, to submit a set of form items to the server you might have:

<submit next="log_request" method="post"
 namelist="name rank serial_number" 
 fetchtimeout="100s" fetchaudio="audio/brahms2.wav"/>

Attributes of <submit> include:

If an ECMAScript object o is the target of a submit then all its (ECMAScript) fields f1, f2, ... are submitted using the names o.f1, o.f2, etc.

If a <submit> contains a variable which references recorded audio but does not contain an ENCTYPE of multipart/form-data, the behavior is not specified. It is probably inappropriate to attempt to URL-encode large quantities of data.

5.3.9 EXIT

Returns control to the interpreter context which determines what to do next.

<exit/>

This element differs from <return> in that it terminates all loaded documents, while <return> returns from a <subdialog> invocation. If the <subdialog> caused a new document (or application) to be invoked, then <return> will cause that document to be terminated, but execution will resume after the <subdialog>.

Note that once <exit> returns control to the interpreter context, the interpreter context is free to do as it wishes. It may play a top level menu for the user, drop the call, or transfer the user to an operator, for example.

Attributes include:

The <exit> element does not throw an "exit" event.

5.3.10 RETURN

Return ends execution of a subdialog and returns control and data to a calling dialog. The attributes are:

In returning from a subdialog, an event can be thrown at the invocation point, or data is returned as an ECMAScript object. A return element that is encountered when not executing as a subdialog throws a semantic error. The example below shows an event propagated from a subdialog to its calling dialog when the subdialog fails to obtain a recognizable result. It also shows data returned under normal conditions.

Form with calling dialog

<form> 
  <subdialog name="result" src="#getssn"> 
    <nomatch> 
      <!-- a no match event that is returned by the
          subdialog indicates that a valid social security
          number could not be matched. --> 
      <goto next="http://myservice.example.com/ssn-problems.vxml"/> 
    </nomatch> 

    <filled> 
      <submit namelist="result.ssn" 
       next="http://myservice.example.com/cgi-bin/process"/> 
    </filled> 
  </subdialog> 
</form>

Subdialog to get social security number

<form id="getssn"> 
  <field name="ssn"> 
      <grammar src="http://grammarlib/ssn.grxml" 
        type="application/grammar+xml"/>
      <prompt> Please say social security number.</prompt> 
      <nomatch count=3> 
        <return event="nomatch"/> 
      </nomatch> 
      <filled> 
        <return namelist="ssn"/> 
      </filled> 
  </field> 
</form>

The subdialog event handler for <nomatch> is triggered on the third failure to match; when triggered, it returns from the subdialog, and includes the nomatch event to be thrown in the context of the calling dialog. In this case, the calling dialog will execute its <nomatch> handler, rather than the <filled> element, where the resulting action is to execute a <goto> element. Under normal conditions, the <filled> element of the subdialog is executed after a recognized social security number is obtained, and then this value is returned to the calling dialog, and is accessible as result.ssn.

5.3.11 DISCONNECT

Causes the interpreter context to disconnect from the user. As a result, the interpreter context will throw a telephone.disconnect.hangup event, which may be caught to do cleanup processing, e.g.

<disconnect/>

A <disconnect> differs from an <exit> in that it forces the interpreter context to drop the call.

5.3.12 SCRIPT

The <script> element allows the specification of a block of client-side scripting language code, and is analogous to the HTML <SCRIPT> element. For example, this document has a script that computes a factorial.

<?xml version="1.0"?> 
<vxml version="2.0"> 
    <script> <![CDATA[ 
       function factorial(n) 
       {
         return (n <= 1)? 1 : n * factorial(n-1);
       } 
    ]]> </script> 

  <form id="form"> 
    <field name="fact" type="number"> 
      <prompt>
        Tell me a number and I'll tell you its factorial.
      </prompt> 
      <filled> 
        <prompt> 
          <value expr="fact"/> factorial is 
          <value expr="factorial(fact)"/> 
        </prompt> 
      </filled> 
    </field> 
  </form> 
</vxml>

A <script> element may occur in the <vxml> and <form> elements, or in executable content (in <filled>, <if>, <block>, <catch>, or the short forms of <catch>). Scripts in the <vxml> element are evaluated just after the document is loaded, along with the <var> elements, in document order. Scripts in the <form> element are evaluated once in document order, along with <var> elements and form item variables, when execution moves into the <form> element. A <script> element in executable content is executed, like other executable elements, as it is encountered.

The <script> element has the following attributes:

The VoiceXML <script> element (unlike the HTML <script> element) does not have a type attribute; ECMAScript is the required scripting language for VoiceXML. Each <script> element is executed in the scope of its containing element; i.e., it does not have its own scope.

Here is a time-telling service with a block containing a script that initializes time variables in the dialog scope of a form:

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <form> 
    <var name="hours"/> 
    <var name="minutes"/> 
    <var name="seconds"/> 
    <block> 
      <script> 
        var d = new Date(); 
        hours = d.getHours(); 
        minutes = d.getMinutes(); 
        seconds = d.getSeconds(); 
      </script> 
    </block> 
    <field name="hear_another" type="boolean"> 
      <prompt> 
        The time is <value expr="hours"/> hours, 
        <value expr="minutes"/> minutes, and 
        <value expr="seconds"/> seconds. 
      </prompt> 
      <prompt>Do you want to hear another time?</prompt> 
      <filled> 
        <if cond="hear_another"> 
          <clear/> 
        </if> 
      </filled> 
    </field> 
  </form> 
</vxml>

The content of a <script> element is evaluated in the same scope as a <var> element (see 5.1.2 Variable Scopes and 5.3.1 VAR).

The ECMAScript scope chain (see section 10.1.4 in http://www.ecma.ch/ecma1/STAND/ECMA-262.htm) is set up so that variables declared either with <var> or inside <script> are put into the scope associated with the element in which the <var> or <script> element occurs. For example, the variable declared in a <script> element under a <form> element has a dialog scope, and can be accessed as a dialog scope variable as follows:

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <form> 
    <script> 
      var now = new Date(); <!-- this has a dialog scope-->
    </script>
    <var name="seconds" expr="now.getSeconds()"/> <!-- this has a dialog scope-->
    <block> 
      <var name="now" expr="new Date()"/> <!-- this has an anonymous scope -->
      <script> 
        var current = now.getSeconds();       <!-- "now" in the anonymous scope -->
        var approx = dialog.now.getSeconds(); <!-- "now" in the dialog scope -->
      </script> 
    </block> 
  </form> 
</vxml>

All variables must be declared before being referenced by ECMAScript scripts, or by VoiceXML elements.

5.3.13 LOG

The <log> element allows an application to generate a debug message which a developer can use to help in application development.

The <log> element may contain any combination of text and <value> elements. The generated message consists of the concatenation of the text and the string form of the value of the "expr" attribute of the <value> elements.

The manner in which the message is displayed or logged is platform-dependent. Platforms are not required to preserve white space or formating controls.

ECMAScript expressions in <log> must be evaluated in document order. The use of the <log> element should have no other side-effects on interpretation.

<log>The card number was <value expr="card_num"/>?</log>

The <log> element has the following attributes:

6. Environment and Resources

6.1 Resource Fetching

6.1.1 Fetching

A VoiceXML interpreter context needs to fetch VoiceXML documents, and other resources, such as audio files, grammars, scripts, and objects. Each fetch of the content associated with a URI is governed by the following attributes:

When content is fetched from a URI, the fetchtimeout attribute determines how long to wait for the content (starting from the time when the resource is needed), and the fetchhint attribute determines when the content is fetched. The caching policy for a VoiceXML interpreter context utilizes the maxage and maxstale attributes and is explained in more detail below.

The fetchhint attribute, in combination with the various fetchhint properties, is merely a hint to the interpreter context about when it may schedule the fetch of a resource.  Telling the interpreter context that it may prefetch a resource does not require that the resource be prefetched; it only suggests that the resource may be prefetched. However, the interpreter context is always required to honor the safe fetchhint.

When transitioning from one dialog to another, through either a <subdialog>, <goto>, <submit>, <link>, or <choice> element, there are additional rules that affect interpreter behavior. If the referenced URI names a document (e.g. "doc#dialog") or query data is provided (through POST or GET), then a new document is obtained (either from a local cache or from a server). When it is obtained, the document goes through its initialization phase (i.e., obtaining and initializing a new application root document if needed, initializing document variables, and executing document scripts). The requested dialog (or first dialog if none is specified) is then initialized and execution of the dialog begins. If the referenced URI names only a fragment (e.g. "#dialog") then no document is obtained, and no initialization of the document is performed. The requested dialog is processed as before.

Elements that fetch VoiceXML documents also support the following additional attribute:

The fetchaudio attribute is useful for enhancing a user experience when there may be noticeable delays while the next document is retrieved. This can be used to play background music, or a series of announcements. When the document is retrieved, the audio file is interrupted if it is still playing.

6.1.2 Caching

The VoiceXML interpreter context, like HTML visual browsers, can use caching to improve performance in fetching documents and other resources; audio recordings (which can be quite large) are as common to VoiceXML documents as images are to HTML pages. In a visual browser it is common to include end user controls to update or refresh content that is perceived to be stale. This is not the case for the VoiceXML interpreter context, since it lacks equivalent end user controls. Thus enforcement of cache refresh is at the discretion of the document through appropriate use of the maxage, and maxstale attributes.

The caching policy used by the VoiceXML interpreter context must adhere to the cache correctness rules of HTTP 1.1 (RFC2616). In particular, the Expires and Cache-Control headers must be honored. The following algorithm summarizes these rules and represents the interpreter context behavior when requesting a resource:

• If the resource is not present in the cache, fetch it from the server using get.
• If the resource is in the cache,
  • If a maxage value is provided,
    • If age of the cached resource <= maxage,
      • If the resource has expired,
        • Perform maxstale check.
      • Otherwise, use the cached copy.
    • Otherwise, fetch it from the server using get.
  • Otherwise,
    • If the resource has expired,
      • Perform maxstale check.
    • Otherwise, use the cached copy.

The "maxstale check" is:

• If maxstale is provided,
  • If cached copy has exceeded its expiration time by no more than maxstale seconds, then use the cached copy.
  • Otherwise, fetch it from the server using get.
• Otherwise, fetch it from the server using get.

Note: it is an optimization to perform a "get if modified" on a document still present in the cache when the policy requires a fetch from the server.

While the maxage and maxstale attributes are drawn from and directly supported by HTTP 1.1, some resources may be addressed by URIs that name protocols other than HTTP. If the protocol does not support the notion of resource age, the interpreter context shall compute a resource's age from the time it was received. If the protocol does not support the notion of resource staleness, the interpreter context shall consider the resource to have expired immediately upon receipt.

6.1.2.1 Controlling the Caching Policy

VoiceXML allows the author to control the caching policy for each use of each resource.

Each resource-related element may specify maxage and maxstale attributes. Setting maxage to a non-zero value can be used to get a fresh copy of a resource that may not have yet expired in the cache. A fresh copy can be unconditionally requested by setting maxage to zero.

Using maxstale enables the author to state that an expired copy of a resource, that is not too stale (according to the rules of HTTP 1.1), may be used. This can improve performance by eliminating a fetch that would otherwise be required to get a fresh copy. It is especially useful for authors who may not have direct server-side control of the expiration dates of large static files.

6.1.3 Prefetching

Prefetching is an optional feature that an interpreter context may implement to obtain a resource before it is needed. A resource that may be prefetched is identified by an element whose fetchhint attribute equals "prefetch". When an interpreter context does prefetch a resource, it must ensure that the resource fetched is precisely the one needed. In particular, if the URI is computed with an expr attribute, the interpreter context must not move the fetch up before any assignments to the expression's variables. Likewise, the fetch for a <submit> must not be moved prior to any assignments of the namelist variables.

The expiration status of a resource must be checked on each use of the resource, and, if its fetchhint attribute is "prefetch", when it is prefetched. The check must follow the caching policy specified in Section 6.1.2.

6.1.4 Protocols

The "http" URI protocol must be supported by VoiceXML platforms, the "https" protocol should be supported and other URI protocols may be supported.

6.2 Meta

The <meta> element specifies meta-data, as in HTML, which is data about the document rather than the document’s content. There are two types of <meta>. The first type specifies a meta-data property of the document as a whole. For example to specify the maintainer of a VoiceXML document:

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <meta name="maintainer" content="jpdoe@anycompany.example.com"/> 
   ... 
</vxml>

The interpreter could use this information, for example, to compose and email an error report to the maintainer.

VoiceXML does not specify required meta-data properties, but the following are recommended:

The second type of <meta> specifies HTTP response headers. In the following example, the first <meta> element sets an expiration date that prevents caching of the document; the second <meta> element sets the Date header.

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <meta http-equiv="Expires" content="0"/> 
  <meta http-equiv="Date"    content="Thu, 12 Dec 2000 23:27:21 GMT"/> 
   ... 
</vxml>

Attributes of <meta> are:

6.3 Property

The <property> element sets a property value. Properties are used to set values that affect platform behavior, such as the recognition process, timeouts, caching policy, etc.

Properties may be defined for the whole application, for the whole document at the <vxml> level, for a particular dialog at the <form> or <menu> level, or for a particular form item. Properties apply to their parent element and all the descendants of the parent. A property at a lower level overrides a property at a higher level. Properties specified in the application root document provide default values for properties in every document in the application; properties specified in an individual document override property values specified in the application root document.

In some cases, <property> elements specify default values for element attributes, such as timeout or bargein. For example, to turn off bargein by default for all the prompts in a particular form:

<form id="no_bargein_form"> 
  <property name="bargein" value="false"/> 
  <block> 
    <prompt>
      This introductory prompt cannot be barged into.
    </prompt> 
    <prompt>
      And neither can this prompt.
    </prompt> 
    <prompt bargein="true">
      But this one <emphasis>can</emphasis> be barged into.
    </prompt> 
  </block> 
   ... 
</form>

6.3.1 Platform-Specific Properties

An interpreter context is free to provide platform-specific properties.  For example, to ensure that one second of silence is prepended in front of each recording made by a particular document: 

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <property name="com.example.acme.endpointing.record_init_silence" value="1s"/> 
   ... dialogs that make recordings go here ... 
</vxml>

Platform-specific properties introduce incompatibilities.  To minimize them, the following interpreter context guidelines are strongly recommended:

• Platform-specific properties should use reverse domain names to eliminate potential collisions: com.acme.output_volume, org.ttscentral.voice.name, etc.

• An interpreter context should not throw an error.unsupported.property event when encountering a property it cannot process; rather the interpreter context should just ignore that property.

6.3.2 Generic Speech Recognizer Properties

The generic speech recognizer properties mostly are taken from the Java Speech API (see
http:// www.javasoft.com/products/java-media/speech/index.html):

6.3.3 Generic DTMF Recognizer Properties

Several generic properties pertain to DTMF grammar recognition:

6.3.4 Prompt and Collect Properties

These properties apply to the fundamental platform prompt and collect cycle:

6.3.5 Fetching Properties

These properties pertain to the fetching of new documents and resources:

6.3.6 Miscellaneous Properties

Our last example shows several of these properties used at multiple levels.

<?xml version="1.0"?> 
<vxml version="2.0"> 
  <!-- set default characteristics for page --> 
  <property name="caching" value="safe"/> 
  <property name="audiofetchhint" value="safe"/> 
  <property name="confidence" value="0.75"/> 

  <form> 
    <!-- override defaults for this form only --> 
    <property name="confidence" value="0.5"/> 
    <property name="bargein" value="false"/> 
    <grammar src="address_book.grxml" type="application/grammar+xml"/>
    <block> 
      <prompt> Welcome to the Voice Address Book </prompt> 
    </block> 
    <initial name="start"> 
      <!-- override default timeout value --> 
      <property name="timeout" value="5s"/> 
      <prompt> Who would you like to call? </prompt> 
    </initial> 
    <field name="person"> 
      <prompt>
        Say the name of the person you would like to call.
      </prompt> 
    </field> 
    <field name="location"> 
      <prompt>
        Say the location of the person you would like to call.
      </prompt> 
    </field> 
    <field name="confirm" type="boolean"> 
      <!-- Use actual utterances to playback recognized words, 
              rather than returned slot values --> 
      <prompt> 
        You said to call <value expr="person$.utterance"/> 
        at <value expr="location$.utterance"/>. 
        Is this correct? 
      </prompt> 
      <filled> 
        <if cond="confirm"> 
          <submit namelist="person location"
            next="http://www.messagecentral.example.com/voice/make_call" />
        </if> 
        <clear/> 
      </filled> 
    </field> 
  </form> 
</vxml>

6.4 Param

The <param> element is used to specify values that are passed to subdialogs or objects. It is modeled on the HTML <PARAM> element. Its attributes are:

Exactly one of expr or value must be present. The use of valuetype and type is optional in general, although they may be required by specific objects. When <param> is contained in a <subdialog> element, the values specified by it are used to initialize dialog <var> elements in the subdialog that is invoked. When <param> is contained in an <object>, the use of the parameter data is specific to the object that is being invoked, and is outside the scope of the VoiceXML specification.

Below is an example of <param> used as part of an <object>. In this case, the first two <param> elements have expressions (implicitly of valuetype="data"), the third <param> has an explicit value, and the fourth is a URI that returns a media type of text/plain. The meaning of this data is specific to the object.

<object name="debit" 
 classid="method://credit_card/gather_and_debit" 
 data="http://www.recordings.example.com/prompts/credit/jesse.jar"/> 
  <param name="amount" expr="document.amt"/> 
  <param name="vendor" expr="vendor_num"/> 
  <param name="application_id" value="ADC5678-QWOO"/> 
  <param name="authentication_server"
   value="http://auth_svr.example.com" 
   valuetype="ref"
   type="text/plain"/> 
</object>

The next example illustrates <param> used with <subdialog>. In this case, two expressions are used to initialize variables in the scope of the subdialog form.

Form with calling dialog

<form> 
  <subdialog name="result" src="http://another.example.com/#getssn"> 
    <param name="firstname" expr="document.first"/> 
    <param name="lastname"  expr="document.last"/> 
    <filled> 
      <submit namelist="result.ssn"
        next="http://myservice.example.com/cgi-bin/process"/> 
    </filled> 
  </subdialog> 
</form>

Subdialog in http://another.example.com

 <form id="getssn"> 
  <var name="firstname"/> 
  <var name="lastname"/> 
  <field name="ssn"> 
      <grammar src="http://grammarlib/ssn.grxml"
        type="application/grammar+xml"/>
      <prompt>
        Please say social security number.
      </prompt> 
      <filled> 
        <if cond="validssn(firstname,lastname,ssn)"> 
          <assign name="status" expr="true"/> 
          <return namelist="status ssn"/> 
        <else/> 
          <assign name="status" expr="false"/> 
          <return namelist="status"/> 
        </if> 
      </filled> 
  </field> 
</form>

Using <param> in a <subdialog> is a convenient way of passing data to a subdialog without requiring the use of server side scripting.

6.5 Time Designations

Time designations follow those used in W3C's Cascading Style Sheet recommendation (http://www.w3.org/TR/REC-CSS2/syndata.html#q20). They consist of an unsigned integer number followed by an optional time unit identifier. The time unit identifiers are:

• ms: milliseconds (the default)

• s: seconds

Examples include: "500", "3s", "850ms", and "+1.5s".  Negative time designations are not permitted.

Appendices

Appendix A — Glossary of Terms

active grammar

A speech or DTMF grammar that is currently active. This is based on the currently executing element, and the scope elements of the currently defined grammars.

application

A collection of VoiceXML documents that are tagged with the same application name attribute.

ASR

Automatic speech recognition.

author

The creator of a VoiceXML document.

catch element

A <catch> block or one of its abbreviated forms. Certain default catch elements are defined by the VoiceXML interpreter.

CSS W3C Cascading Style Sheet specification.

See http://www.w3.org/TR/REC-CSS2/syndata.html#q20

dialog

An interaction with the user specified in a VoiceXML document. Types of dialogs include forms and menus.

ECMAScript

A standard version of JavaScript backed by the European Computer Manufacturer’s Association. See http://www.ecma.ch/ecma1/STAND/ECMA-262.htm

event

A notification "thrown" by the implementation platform, VoiceXML interpreter context, VoiceXML interpreter, or VoiceXML code. Events include exceptional conditions (semantic errors), normal errors (user did not say something recognizable), normal events (user wants to exit), and user defined events.

executable content

Procedural logic that occurs in <block>, <filled>, and event handlers.

field item

A form item whose purpose is to input a field item variable. Field items include <field>, <record>, <object>, <subdialog>, and <transfer>.

form

A dialog that interacts with the user in a highly flexible fashion with the computer and the user sharing the initiative.

form item

An element of <form> that can be visited during form execution: <initial>, <block>, <field>, <record>, <object>, and <transfer>.

form item variable

A variable, either implicitly or explicitly defined, associated with each form item in a form. If the form item variable is undefined, the form interpretation algorithm will visit the form item and use it to interact with the user.

implementation platform

A computer with the requisite software and/or hardware to support the types of interaction defined by VoiceXML.

link

A set of grammars that when matched by something the user says or keys in, either transitions to a new dialog or document or throws an event in the current form item.

menu

A dialog presenting the user with a set of choices and takes action on the selected one.

mixed initiative

A computer-human interaction in which either the computer or the human can take initiative and decide what to do next.

JSGF

Java API Speech Grammar Format. A proposed standard for representing speech grammars. See http://www.javasoft.com/products/java-media/speech/forDevelopers/JSGF

JSML

Java API Speech Markup Language. A proposed standard for speech markups. See http://www.javasoft.com/products/java-media/speech/forDevelopers/JSML

object

A platform-specific capability with an interface available via VoiceXML.

request

A collection of data including: a URI specifying a document server for the data, a set of name-value pairs of data to be processed (optional), and a method of submission for processing (optional).

SABLE

A consortium seeking to develop standards for speech markup. See http://www.bell-labs. com/project/tts/sable.html

script

A fragment of logic written in a client-side scripting language, especially ECMAScript, which is a scripting language that must be supported by any VoiceXML interpreter.

session

A connection between a user and an implementation platform, e.g. a telephone call to a voice response system. One session may involve the interpretation of more than one VoiceXML document.

subdialog

A VoiceXML dialog (or document) invoked from the current dialog in a manner analogous to function calls.

tapered prompts

A set of prompts used to vary a message given to the human. Prompts may be tapered to be more terse with use (field prompting), or more explicit (help prompts).

throw

An element that fires an event.

TTS

Text-To-Speech; speech synthesis.

user

A person whose interaction with an implementation platform is controlled by a VoiceXML interpreter.

URI

Uniform Resource Indicator.

URL

Uniform Resource Locator.

VoiceXML document

An XML document conforming to the VoiceXML specification.

VoiceXML interpreter

A computer program that interprets a VoiceXML document to control an implementation platform for the purpose of conducting an interaction with a user.

VoiceXML interpreter context

A computer program that uses a VoiceXML interpreter to interpret a VoiceXML Document and that may also interact with the implementation platform independently of the VoiceXML interpreter.

W3C

World Wide Web Consortium http://www.w3.org/

Appendix B — VoiceXML Document Type Definition

This section is Normative.

<!-- 

   VoiceXML 2.0 DTD 
   version 20010525
   Copyright (c) 2001 W3C (MIT, INRIA, Keio)
   All Rights Reserved.

   This DTD is the merger of 
    VoiceXML 1.0
    Speech Synthesis Markup Language 1.0
    Speech Recognition Grammar Format
   with changes, additions, deletions to each noted.

-->

<!ENTITY % audio 
    "#PCDATA | audio | enumerate | value" >


<!ENTITY % bargeintype   "(energy | speech | recognition)" >


<!ENTITY % boolean   "(true|false)" >

<!ENTITY % content.type "CDATA">

<!ENTITY % duration "CDATA" >

<!ENTITY % event.handler "catch | help | noinput | nomatch | error" >

<!ENTITY % event.name "NMTOKEN" >

<!ENTITY % event.names "CDATA" >

<!ENTITY % executable.content 
    "%audio; | assign | clear | disconnect | exit | goto | if | log | prompt | 
    reprompt | return | script | submit | throw | var " >


<!ENTITY % expression "CDATA" >

<!ENTITY % field.name "NMTOKEN" >


<!ENTITY % field.names "CDATA" >


<!ENTITY % integer "CDATA" >

<!ENTITY % item.attrs
    "name       %field.name;    #IMPLIED 
    cond        %expression;    #IMPLIED
    expr        %expression;    #IMPLIED " >

<!ENTITY % uri "CDATA" >


<!ENTITY % cache.attrs
    "fetchhint  (prefetch|safe)  #IMPLIED
    fetchtimeout    %duration;  #IMPLIED
    maxage      %integer;   #IMPLIED
    maxstale    %integer;   #IMPLIED" >


<!ENTITY % next.attrs
    "next       %uri;       #IMPLIED
    expr        %expression;    #IMPLIED " >

<!ENTITY % submit.attrs
    "method     (get|post)  'get'
    enctype     %content.type;  'application/x-www-form-urlencoded'
    namelist    %field.names;   #IMPLIED" >


<!ENTITY % sentence-elements 
    "break | emphasis | mark | phoneme | prosody | say-as | voice" >



<!ENTITY % allowed-within-sentence " %audio; | %sentence-elements; " >



<!ENTITY % structure "paragraph | p | sentence | s">

<!ENTITY % phoneme-string "CDATA" >
<!ENTITY % phoneme-alphabet "CDATA" >



<!ENTITY % tts       "%sentence-elements; | %structure;" >


<!ENTITY % variable  "block | field | var" >

<!--================================= Root ================================-->


<!ELEMENT vxml       
    (%event.handler; | form | link | menu | meta | 
    property | script | var)+ >
<!ATTLIST vxml 
    application %uri;       #IMPLIED
    base        %uri;       #IMPLIED
    xml:lang    NMTOKEN     #IMPLIED
    version     CDATA       #REQUIRED >


<!ELEMENT meta       EMPTY >
<!ATTLIST meta
    name        NMTOKEN     #IMPLIED
    content     CDATA       #REQUIRED 
    http-equiv  NMTOKEN     #IMPLIED >

<!--================================= Dialogs =============================-->


<!ENTITY % input "grammar" >


<!ENTITY % scope     "(document | dialog)" >


<!ELEMENT form
    (%input; | %event.handler; | filled | initial | object | link | property |
    record | script | subdialog | transfer | %variable;)* >
<!ATTLIST form
    id      ID      #IMPLIED 
    scope       %scope;     'dialog' >



<!ENTITY % accept.attrs
    "accept     (exact | approximate)   'exact'" >



<!ELEMENT menu       
    (%audio; | choice | %event.handler; | prompt | property)* >
<!ATTLIST menu 
    id      ID      #IMPLIED 
    scope       %scope;     'dialog'
    %accept.attrs;
    dtmf        %boolean;   'false' >



<!ELEMENT choice (%audio; | grammar | %tts;)* >
<!ATTLIST choice
    %cache.attrs;
    %accept.attrs;
    dtmf        CDATA       #IMPLIED
    event       %event.name;    #IMPLIED
    fetchaudio  %uri;       #IMPLIED
    %next.attrs; >


<!--================================ Prompts ==============================-->


<!ELEMENT prompt (%audio; | %tts;)* >
<!ATTLIST prompt
    bargein     %boolean;   #IMPLIED
    bargeintype %bargeintype;   #IMPLIED
    cond        %expression;    #IMPLIED
    count       %integer;   #IMPLIED
    xml:lang    NMTOKEN     #IMPLIED
    timeout     %duration;  #IMPLIED >


<!ELEMENT enumerate  (%audio; | %tts;)*>

<!ELEMENT reprompt   EMPTY >

<!--================================ Fields ===============================-->

<!ELEMENT field      
    (%audio; | %event.handler; | filled | %input; | link | option | prompt | property)* >
<!ATTLIST field
    %item.attrs; 
    type        CDATA       #IMPLIED
    slot        NMTOKEN     #IMPLIED 
    modal       %boolean;   'false' >

<!ELEMENT option     (#PCDATA) >
<!ATTLIST option
    dtmf        CDATA       #IMPLIED
    value       CDATA       #IMPLIED >

<!ELEMENT var        EMPTY >
<!ATTLIST var
    name        %field.name;    #REQUIRED
    expr        %expression;    #IMPLIED >

<!ELEMENT initial    (%audio; | %event.handler; | link | prompt | property)* >
<!ATTLIST initial
    %item.attrs; >

<!ELEMENT block  (%executable.content;)* >
<!ATTLIST block
    %item.attrs; >

<!ELEMENT assign EMPTY >
<!ATTLIST assign
    name        %field.name;    #REQUIRED
    expr        %expression;    #REQUIRED >

<!ELEMENT clear  EMPTY >
<!ATTLIST clear
    namelist    %field.names;   #IMPLIED >


<!ELEMENT value  EMPTY >
<!ATTLIST value
    expr        %expression;    #REQUIRED >


<!--================================== Events =============================-->

<!ENTITY % event.handler.attrs 
    "count      %integer;   #IMPLIED
    cond        %expression;    #IMPLIED" >

<!ELEMENT catch  (%executable.content;)* >
<!ATTLIST catch
    event       %event.names;   #REQUIRED   
    %event.handler.attrs; >

<!ELEMENT error  (%executable.content;)* >
<!ATTLIST error
    %event.handler.attrs; >

<!ELEMENT help       (%executable.content;)* >
<!ATTLIST help
    %event.handler.attrs; >


<!ELEMENT link       (grammar)* >
<!ATTLIST link
    %cache.attrs;
    %next.attrs; 
    fetchaudio  %uri;       #IMPLIED
    dtmf        CDATA       #IMPLIED
    event       %event.name;    #IMPLIED >


<!ELEMENT noinput    (%executable.content;)* >
<!ATTLIST noinput
    %event.handler.attrs; >

<!ELEMENT nomatch    (%executable.content;)* >
<!ATTLIST nomatch
    %event.handler.attrs; >


<!ELEMENT throw  EMPTY >
<!ATTLIST throw
    event       %event.name;    #IMPLIED
    eventexpr   %expression;    #IMPLIED
    message CDATA           #IMPLIED
    messageexpr %expression;    #IMPLIED >

<!--============================== Audio Output ===========================-->


<!ELEMENT audio  (%audio; | %tts;)* >
<!ATTLIST audio
    src     %uri;       #IMPLIED 
    expr        %expression;    #IMPLIED
    %cache.attrs; >

<!ELEMENT paragraph (%allowed-within-sentence; | sentence | s)*>
<!ATTLIST paragraph
    xml:lang    NMTOKEN     #IMPLIED >

<!ELEMENT p (%allowed-within-sentence; | sentence | s)*>
<!ATTLIST p
        xml:lang        NMTOKEN         #IMPLIED >

<!ELEMENT sentence (%allowed-within-sentence;)*>
<!ATTLIST sentence
    xml:lang    NMTOKEN     #IMPLIED >

<!ELEMENT s (%allowed-within-sentence;)*>
<!ATTLIST s
        xml:lang        NMTOKEN         #IMPLIED >

<!-- The flexible container elements can occur within paragraph -->
<!-- and sentence but may also contain these structural elements. -->

<!ENTITY % voice-name "CDATA">

<!ELEMENT voice (%allowed-within-sentence; | %structure;)*>
<!ATTLIST voice
     gender   (male|female|neutral)                  #IMPLIED
     age      %integer;                              #IMPLIED
                 category (child|teenager|adult|elder)           #IMPLIED
     variant  %integer;                              #IMPLIED
     name     CDATA                                  #IMPLIED >

<!ELEMENT prosody (%allowed-within-sentence; | %structure;)*>
<!ATTLIST prosody
     pitch      CDATA  #IMPLIED
     contour    CDATA  #IMPLIED
     range      CDATA  #IMPLIED
     rate       CDATA  #IMPLIED
     duration   CDATA  #IMPLIED
     volume     CDATA  #IMPLIED >

<!-- These basic container elements can contain any of the -->
<!-- within-sentence elements, but neither sentence or paragraph. -->

<!ELEMENT emphasis (%allowed-within-sentence;)*>
<!ATTLIST emphasis
     level      (strong|moderate|none|reduced)  'moderate' >

<!-- This basic container element can contain data or value -->

<!ELEMENT say-as (#PCDATA|value)>
<!ATTLIST say-as
     type   CDATA            #REQUIRED
     sub    CDATA            #IMPLIED >

<!-- This basic container element can contain only data -->

<!ELEMENT phoneme (#PCDATA)>
<!ATTLIST phoneme
     ph        %phoneme-string;   #REQUIRED
     alphabet  %phoneme-alphabet; #IMPLIED >

<!-- Definitions of the basic empty elements -->

<!ELEMENT break EMPTY>
<!ATTLIST break
     size      (large|medium|small|none)  'medium'
     time      %duration;                 #IMPLIED >

<!ELEMENT mark EMPTY>
<!ATTLIST mark
     name      CDATA   #REQUIRED >


<!--============================= Audio Input =============================-->

<!ENTITY % key   "CDATA" >


<!ENTITY % grammar.attrs
    "%cache.attrs;
    version     CDATA       '1.0'
    xml:lang    NMTOKEN     #IMPLIED
    root        IDREF       #IMPLIED
    mode        (voice | dtmf)  'voice'
    scope       %scope;     #IMPLIED
    src     %uri;       #IMPLIED 
    type        CDATA       #IMPLIED
    weight      CDATA       #IMPLIED " >


<!ENTITY % rule-expansion "#PCDATA | token | ruleref
                              | item | one-of | count " >

<!ELEMENT ruleref EMPTY>
<!ATTLIST ruleref
     uri CDATA #IMPLIED
     import CDATA #IMPLIED
     special CDATA #IMPLIED
     xml:lang NMTOKEN #IMPLIED
     tag CDATA #IMPLIED>

<!ELEMENT token (#PCDATA)>
<!ATTLIST token
     lexicon CDATA #IMPLIED
     xml:lang NMTOKEN #IMPLIED>

<!ELEMENT one-of (item)*>
<!ATTLIST one-of
     tag CDATA #IMPLIED
     xml:lang NMTOKEN #IMPLIED>

<!ELEMENT item ( %rule-expansion; )*>
<!ATTLIST item
    weight NMTOKEN #IMPLIED
    tag CDATA #IMPLIED
    xml:lang NMTOKEN #IMPLIED>

<!ELEMENT count ( %rule-expansion; )*>
<!ATTLIST count
    number CDATA #IMPLIED
    tag CDATA #IMPLIED
    xml:lang NMTOKEN #IMPLIED>

<!ELEMENT rule ( %rule-expansion; | example )*>
<!ATTLIST rule
    id ID #REQUIRED
    scope (private | public) "private">

<!ELEMENT example (#PCDATA)>

<!ELEMENT import EMPTY>
<!ATTLIST import
    uri CDATA #REQUIRED
    name CDATA #REQUIRED>

<!ELEMENT grammar    (#PCDATA | import | rule )* >
<!ATTLIST grammar
    %grammar.attrs; >


<!ELEMENT record 
    (%audio; | %event.handler; | filled | grammar | prompt | property)* >    
<!ATTLIST record
    %item.attrs;
    type        CDATA       #IMPLIED
    beep        %boolean;   'false'
    dest        %uri;       #IMPLIED
    maxtime     %duration;  #IMPLIED 
    modal       %boolean;   'true'
    finalsilence    %duration;  #IMPLIED
    dtmfterm    %boolean;   'true'  >


<!--============================ Call Control ============================-->

<!ELEMENT disconnect EMPTY >

<!ELEMENT transfer   
    (%audio; | %event.handler;  | filled | grammar | prompt | property)* >
<!ATTLIST transfer
    %item.attrs;
    dest        %uri;       #IMPLIED
    destexpr    %expression;    #IMPLIED
    bridge      %boolean;   'false' 
    connecttimeout  %duration;  #IMPLIED
    maxtime     %duration;  #IMPLIED
    transferaudio   %uri;       #IMPLIED >

<!--============================ Control Flow ============================-->

<!ENTITY % if.attrs 
    "cond       %expression;    #REQUIRED" >

<!ELEMENT if         (%executable.content; | elseif | else)* >
<!ATTLIST if
    %if.attrs; >

<!ELEMENT elseif EMPTY >
<!ATTLIST elseif 
    %if.attrs; >

<!ELEMENT else   EMPTY >

<!ELEMENT exit       EMPTY >
<!ATTLIST exit
    expr        %expression;    #IMPLIED 
    namelist    %field.names;   #IMPLIED >

<!ELEMENT filled     (%executable.content;)* >
<!ATTLIST filled
    mode        (any|all)   "all"
    namelist    %field.names;   #IMPLIED >

<!ELEMENT goto       EMPTY >
<!ATTLIST goto
    %cache.attrs;
    %next.attrs;
    fetchaudio  %uri;       #IMPLIED
    expritem    %expression;    #IMPLIED 
    nextitem    %field.name;    #IMPLIED >

<!ELEMENT param  EMPTY >
<!ATTLIST param
    name        NMTOKEN     #REQUIRED
    expr        %expression;    #IMPLIED
    value       CDATA       #IMPLIED
    valuetype   (data|ref)  'data'
    type        CDATA       #IMPLIED >

<!ELEMENT return EMPTY >
<!ATTLIST return
    namelist    %field.names;   #IMPLIED
    event       %event.name;    #IMPLIED >

<!ELEMENT subdialog  
    (%audio; | %event.handler; | filled | param | prompt | property)* >
<!ATTLIST subdialog
    %item.attrs;
    src     %uri;       #REQUIRED
    %cache.attrs;
    fetchaudio  %uri;       #IMPLIED
    %submit.attrs; >

<!ELEMENT submit EMPTY >
<!ATTLIST submit
    %cache.attrs;
    %next.attrs;
    fetchaudio  %uri;       #IMPLIED
    %submit.attrs; >

<!--========================== Miscellaneous ==============================-->


<!ELEMENT log        (#PCDATA | value)* >
<!ATTLIST log
    label       CDATA       #IMPLIED
    expr        %expression;    #IMPLIED >


<!ELEMENT object 
    (%audio; | %event.handler; | filled | param | prompt | property)* >
<!ATTLIST object
    %item.attrs;
    %cache.attrs;
    classid     %uri;       #IMPLIED
    codebase    %uri;       #IMPLIED
    data        %uri;       #IMPLIED
    type        CDATA       #IMPLIED
    codetype    CDATA       #IMPLIED
    archive     %uri;       #IMPLIED >

<!ELEMENT property   EMPTY >
<!ATTLIST property
    name        NMTOKEN     #REQUIRED
    value       CDATA       #REQUIRED >

<!ELEMENT script (#PCDATA) >
<!ATTLIST script
    src     %uri;       #IMPLIED
    charset     CDATA       #IMPLIED
    %cache.attrs; >

Appendix C — Form Interpretation Algorithm

The form interpretation algorithm (FIA) drives the interaction between the user and a VoiceXML form or menu. A menu can be viewed as a form containing a single field whose grammar and whose <filled> action are constructed from the <choice> elements.

The FIA must handle:

• Form initialization.

• Prompting, including the management of the prompt counters needed for prompt tapering.

• Grammar activation and deactivation at the form and form item levels.

• Entering the form with an utterance that matched one of the form’s document-scoped grammars while the user was visiting a different form or menu.

• Leaving the form because the user matched another form, menu, or link’s document-scoped grammar.

• Processing multiple field fills from one utterance, including the execution of the relevant <filled> actions.

• Selecting the next form item to visit, and then processing that form item.

• Choosing the correct catch element to handle any events thrown while processing a form item.

First we define some terms and data structures used in the form interpretation algorithm:

active grammar set

The set of grammars active during a VoiceXML interpreter context’s input collection operation.

utterance

A summary of what the user said or keyed in, including the specific grammar matched, and a dictionary of slot name/slot value pairs. An example utterance might be: "grammar 123 was matched, and the slots are from_city = ‘chicago’, to_city = ‘new orleans’, and flight_num = 2233".

execute

To execute executable content – either a block, a filled action, or a set of filled actions. If an event is thrown during execution, the execution of the executable content is aborted. The appropriate event handler is then executed, and this may cause control to resume in a form item, in the next iteration of the form’s main loop, or outside of the form. If a <goto> is executed, the transfer takes place immediately, and the remaining executable content is not executed.

Here is the conceptual form interpretation algorithm. The FIA can start with no initial utterance, or with an initial utterance passed in from another dialog:

//
// Initialization Phase
//

foreach ( <var> and form item variable, in document order )
   Declare the variable, initializing it to the value of
   the "expr" attribute, if any, or else to undefined.

foreach ( field item )
   Declare a prompt counter and set it to 1.

if ( there is an initial item )
   Declare a prompt counter and set it to 1.

if ( user entered form by speaking to its
     grammar while in a different form )
{
   Enter the main loop below, but start in
   the process phase, not the select phase:
   we already have a collection to process.
}

//
// Main Loop: select next form item and execute it.
//

while ( true )
{
   //
   // Select Phase: choose a form item to visit.
   //

   if ( the last main loop iteration ended
             with a <goto nextitem> )
       Select that next form item.

   else if (there is a form item with an
             unsatisfied guard condition )
       Select the first such form item in document order.

   else
       Do an <exit/> -- the form is full and specified no transition.

   //
   // Collect Phase: execute the selected form item.
   //
   // Queue up prompts for the form item.

   unless ( the last loop iteration ended with
            a catch that had no <reprompt> )
   {
       Select the appropriate prompts for the form item.

       Queue the selected prompts for play prior to
       the next collect operation.

       Increment the form item’s prompt counter.
   }

   // Activate grammars for the form item.

   if ( the form item is modal )
       Set the active grammar set to the form item grammars,
       if any. (Note that some form items, e.g. <block>,
       cannot have any grammars).
   else
       Set the active grammar set to the form item
       grammars and any grammars scoped to the form,
       the current document, the application root
       document, and then elements up the <subdialog>
       call chain.

   // Execute the form item.

   if ( a <field> was selected )
       Collect an utterance or an event from the user.
   else if ( a <record> was chosen )
       Collect an utterance (with a name/value pair
       for the recorded bytes) or event from the user.
   else if ( an <object> was chosen )
       Execute the object, setting the <object>’s
       form item variable to the returned ECMAScript value.
   else if ( a <subdialog> was chosen )
       Execute the subdialog, setting the <subdialog>’s
       form item variable to the returned ECMAScript value.
   else if ( a <transfer> was chosen )
       Do the transfer, and (if wait is true) set the
       <transfer> form item variable to the returned
       result status indicator.
   else if ( the <initial> was chosen )
       Collect an utterance or an event from the user.
   else if ( a <block> was chosen )
   {
       Set the block’s form item variable to a defined value.

       Execute the block’s executable context.
   }

   //
   // Process Phase: process the resulting utterance or event.
   //

   // Process an event.

   if ( the form item execution resulted in an event )
   {
       Find the appropriate catch for the event 
       starting in the scope of the current form item.
       Execute the catch (this may leave the FIA).

       continue
   }

   // Must have an utterance: process ones from outside grammars.

   if ( the utterance matched a grammar from outside the form )
   {
       if ( the grammar belongs to a <link> element )
          Execute that link’s goto or throw.

       if ( the grammar belongs to a menu’s <choice> element )
          Execute the choice’s goto or throw, leaving the FIA.

       // The grammar belongs to another form (or menu).

       Transition to that form (or menu), carrying the utterance
       to the other form (or menu)’s FIA.
   }

   // Process an utterance spoken to a grammar from this form.
   // First copy utterance slot values into corresponding
   // form item variables.

   Clear all "just_filled" flags.

   foreach ( slot in the user’s utterance )
   {
       if ( the slot corresponds to a field item )
       {
          Copy the slot value into the field item’s
          form item variable.

          Set this field item’s "just_filled" flag.
       }
   }

   // Set <initial> form item variable if any field items are filled.

   if ( any field item variable is set as a result of the user utterance )
       Set the <initial> form item variable.

   // Next execute any <filled> actions triggered by this utterance.

   foreach ( <filled> action in document order )
   {
       // Determine the form item variables the <filled> applies to.

       N = the <filled>’s "namelist" attribute.

       if ( N equals "" )
       {
          if ( the <filled> is a child of a form item )
            N = the form item’s form item variable name.
          else if ( the <filled> is a child of a form )
            N = the form item variable names of all the field
                items in that form.
       }

       // Is the <filled> triggered?

       if ( any form item variable in the set N was "just_filled"
              AND  (  the <filled> mode is "all"
                          AND all variables in N are filled
                      OR the <filled> mode is "any"
                          AND any variables in N are filled) )
            Execute the <filled> action.

        If an event is thrown during the execution of a <filled>, 
            event handler selection starts in the scope of the <filled>,
        which could be a form item or the form itself.
   }
}

Appendix D — Timing Properties

The various timing properties for speech and DTMF recognition work together to define the user experience. The ways in which these different timing parameters function are outlined in the timing diagrams below. In these diagrams, the start for wait of DTMF input, or user speech both occur at the time that the last prompt has finished playing.

A.1. DTMF Grammars

DTMF grammars use timeout, interdigittimeout, termtimeout and termchar to tailor the user experience. The effects of these are shown in the following timing diagrams.

timeout, No Input Provided

The timeout parameter determines when the <noinput> event is thrown because the user has failed to enter any DTMF (Figure 8).

Figure 8: Timing diagram for timeout when no input provided.

interdigittimeout, Grammar is Not Ready to Terminate

In Figure 9, the interdigittimeout determines when the nomatch event is thrown because a DTMF grammar is not yet recognized, and the user has failed to enter additional DTMF.

Figure 9: Timing diagram for interdigittimeout, grammar is not ready to terminate.

interdigittimeout, Grammar is Ready to Terminate

The example below shows the situation when a DTMF grammar could terminate, or extend by the addition of more DTMF input, and the user has elected not to provide any further input.

Figure 10: Timing diagram for interdigittimeout, grammar is ready to terminate.

termchar and interdigittimeout, Grammar Can Terminate

In the example below, a termchar is non-empty, and is entered by the user before an interdigittimeout expires, to signify that the users DTMF input is complete; the termchar is not included as part of the recognized value.

Figure 11: Timing diagram for termchar and interdigittimeout, grammar can terminate.

termchar Empty When Grammar Must Terminate

In the example below, the entry of the last DTMF has brought the grammar to a termination point at which no additional DTMF is expected. Since termchar is empty, there is no optional terminating character permitted, thus the recognition ends and the recognized value is returned.

Figure 12:Timing diagram for termchar empty when grammar must terminate.

termchar Non-Empty and termtimeout When Grammar Must Terminate

In the example below, the entry of the last DTMF has brought the grammar to a termination point at which no additional DTMF is allowed by the grammar. If the termchar is non-empty, then the user can enter an optional termchar DTMF. If the user fails to enter this optional DTMF within termtimeout, the recognition ends and the recognized value is returned. If the termtimeout is 0s (the default), then the recognized value is returned immediately after the last DTMF allowed by the grammar, without waiting for the optional termchar.

Figure 13: Timing diagram for termchar non-empty and termtimeout when grammar must terminate.

termchar Non-Empty and termtimeout When Grammar Must Terminate

In this last DTMF example, the entry of the last DTMF has brought the grammar to a termination point at which no additional DTMF is allowed by the grammar. Since the termchar is non-empty, the user enters the optional termchar within termtimeout causing the recognized value to be returned (excluding the termchar).

Figure 14: Timing diagram for termchar non-empty when grammar must terminate.

A.2. Speech Grammars.

Speech grammars use timeout, completetimeout, and incompletetimeout to tailor the user experience. The effects of these are shown in the following timing diagrams.

timeout When No Speech Provided

In the example below, the timeout parameter determines when the noinput event is thrown because the user has failed to speak.

Figure 15: Timing diagram for timeout when no speech provided.

completetimeout With Speech Grammar Recognized

In the example above, the user provided a utterance that was recognized by the speech grammar. After a silence period of completetimeout has elapsed, the recognized value is returned.

Figure 16: Timing diagram for completetimeout with speech grammar recognized.

incompletetimeout with Speech Grammar Unrecognized

In the example above, the user provided a utterance that is not as yet recognized by the speech grammar but is the prefix of a legal utterance. After a silence period of incompletetimeout has elapsed, a nomatch event is thrown.

Figure 17: Timing diagram for incompletetimeout with speech grammar unrecognized.

Appendix E — Audio File Formats

VoiceXML requires that a platform support the playing and recording audio formats specified below.

The 'audio/basic' mime type is commonly used with the 'au' header format as well as the headerless 8-bit 8Khz mu-law format. If this mime type is specified for recording, the mu-law format must be used. For playback with the 'audio/basic' mime type, platforms must support the mu-law format and may support the 'au' format.

Appendix F —Conformance

This section is Normative.

Conforming VoiceXML Document

A document is a Conforming VoiceXML Document if:

• it is an XML document.
• (relative to XML) is well-formed.
• conforms to the following W3C Recommendations:
  • the XML 1.0 specification (Extensible Markup Language (XML) 1.0).
  • (if any namespaces other than vxml are used in the document) Namespaces in XML.
• its root element is a vxml element. The root element of the document designates the vxml namespace using the "xmlns" attribute (Namespaces in XML). The namespace for vxml is defined to be "http://www.w3.org/2001/vxml".
• if all non-VoiceXML namespace elements and attributes and all xmlns attributes which refer to non-VoiceXML namespace elements are removed from the given document, and if an appropriate XML declaration (i.e., <?xml...?>) is included at the top of the document, and if an appropriate document type declaration (i.e., <!DOCTYPE vxml ... >) which points to the VoiceXML DTD is included immediately thereafter, the result is a valid XML document.

The VoiceXML language or these conformance criteria provide no designated size limits on any aspect of VoiceXML documents. There are no maximum values on the number of elements, the amount of character data, or the number of characters in attribute values.

Conforming VoiceXML Processors

A VoiceXML processor is a program that can parse and process Conforming VoiceXML documents.

In a Conforming VoiceXML Processor, the XML parser must be able to parse and process all well-formed XML constructs defined within XML 1.0 and XML Namespaces. It is not required that a Conforming VoiceXML processor use a validating parser.

A Conforming VoiceXML Processor must be a Conforming Speech Synthesis Markup Language Processor and a Conforming XML Grammar Processor except for differences described in this document. If a syntax error is detected processing a grammar document, then an "error.badfetch" event must be thrown.

A Conforming VoiceXML Processor must support the syntax and semantics of all VoiceXML elements as described in this document. Consequently, a Conforming VoiceXML Processor must not throw an 'error.unsupported.<element>' for any VoiceXML element which must be supported when processing a Conforming VoiceXML Document.

When a Conforming VoiceXML Processor encounters a Conforming VoiceXML Document with non-VoiceXML elements or attributes which are propriety, defined only in earlier versions of VoiceXML, or defined in a non-VoiceXML namespace, and which cannot be processed, then it must throw an "error.unsupported.<element>" event.

There is, however, no conformance requirement with respect to performance characteristics of the VoiceXML Processor.

Determination of whether a given platform is a Conforming VoiceXML Processor will be carried out by the VoiceXML Forum.

Appendix G —Internationalization

VoiceXML is an application of XML 1.0 and thus supports Unicode which defines a standard universal character set.

Additionally, VoiceXML provides a mechanism for precise control of the input and output languages via the use of "xml:lang" attribute. This facility provides:

• the ability to specify the input and output language overriding the VoiceXML Processor default language
• The ability to produce multi-language output
• The ability to interpret input in a language different from the output language(s)

Appendix H —Accessibility

This appendix explains how accessibility guidelines published by W3C's Web Accessibility Initiative (WAI) apply to VoiceXML.

1. The Web Content Accessibility Guidelines 1.0 explains how authors can create Web content that is accessible to people with disabilities. VoiceXML only addresses content which is aimed at the aural modality. To support content accessibility, a service developer should provide the same content via other channels which support other modalities; for example, HTML, WML, etc.
2. The Authoring Tool Accessibility Guidelines 1.0 explains how developers can design accessible authoring tools. VoiceXML authoring tools are not addressed in this specification.
3. The User Agent Accessibility Guidelines 1.0 explains how developers can design accessible user agents. Since VoiceXML explicitly addresses the aural modality and DTMF input, VoiceXML User Agents will not conform to the UAAG.

Appendix I —Privacy

A future revision of this document may specify criteria by which a VoiceXML Processor safeguards the privacy of personal data.

Appendix J —Changes from VoiceXML 1.0

The following is a summary of the differences between VoiceXML 2.0 and VoiceXML 1.0.

Developers of VoiceXML 1.0 applications should pay particular attentions to the changes incompatible with VoiceXML 1.0 specified in Obsolete Elements and Incompatibly Modified Elements.

New Elements

• <log> to specify a debug message (5.3.13)

Obsolete Elements

• <dtmf> superceded by <grammar> with "mode=dtmf" (3.1.2)
• <emp>, <div>, <pros>, and <sayas> JSML elements have been replaced by corresponding elements in Speech Synthesis Markup Language (4.1.2)

Incompatibly Modified Elements

• changed "lang" to "xml:lang" in <vxml> (1.5.1).
• added "accept" attribute to <menu> and <choice> (2.2)
• removed "modal" attribute of <subdialog> (2.3.4)
• removed "fetchaudio" attribute from <object>(2.3.5)
• removed notational equivalence between <filled> in a field and a form-level <filled> triggering on that field (2.4)
• TTS content in <choice>, <prompt>, <enumerate>, and <audio> replaced with definition in Speech Synthesis Markup Language (4.1.2)
• removed "class", "mode" and "recsrc" attributes from <value> (4.1.4)
• changed standard session variable "session.uui" to "session.telephone.uui" (5.1.4)
• removed "caching" attribute (6.1)
• added "maxage" and "maxstale" attributes (6.1)
• removed 'stream' as value of fetchhint property (6.1.1, 6.3.5)
• removed "caching" from fetching properties (6.3.5).

Modified Elements

• added "dest" attribute to <record>; added "maxtime" and "dest" shadow variables (2.3.6)
• added "transferaudio" attribute to <transfer>; added "maxtimedisconnect", and "unknown" as values of bridge transfers; added more error.telephone events (2.3.7)
• added 'dtmf' attribute to <link> (2.5).
• the XML Form of the W3C Speech Recognition Grammar Format, must be supported in <grammar> (3.1).
• added "weight", "mode", "xml:lang", "root" and "version" attributes to <grammar> (3.1).
• added "xml:lang" attribute to <prompt> (4.1).
• added "bargeintype" attribute to <prompt> (4.1).
• added "expr" attribute to <audio> element (4.1.3)
• "bargeintype" attribute on <prompt> (4.1.3)
• added standard session variables "session.telephone.rdnis" and "session.telephone.redirect_reason" (5.1.4)
• added application variable "application.lastresult$" describing last recognition result, including n-best (5.1.5).
• added "event", "eventexpr", "message" and "messageexpr" as attributes of <throw> (5.2.1).
• added "_event" variable to <catch> (5.2.2).
• added "error.badfetch.http.nnn" as pre-defined error type (5.2.6)
• added "errrobadfetch.protocol.<response code>" as pre-defined error type (5.2.6)
• added "maxspeechtimeout" event (5.2.6)
• added "error.unsupported.language" pre-defined error type (5.3.6)
• added required support for "multipart/form-data" value as "enctype" of <submit> (5.3.8)
• <script> can occur in <form> element (5.3.12)
• HTTP is mandatory (6.1.4)
• added "maxspeechtimeout" property (6.3.2)
• added "bargeintype" property (6.3.4)
• added "fetchaudiodelay" and "fetchaudiominimum" to fetch properties (6.3.5)
• added "maxnbest" session property (6.3.6)
• added "universals" property (with default "none") (6.3.6).
• Added defaults for fetching attributes, as well as "maxage" and "maxstale" fetching attributes to: <choice>, <subdialog>, <object>, <link>, <grammar>, <audio>, <goto>, <submit> and <script>

Clarifications

• definitions of, and transitions between, root and leaf document (1.5.2).
• referencing application root document and its grammars (1.5.2).
• <subdialog> transferring control to another <subdialog> and to another dialog via <goto> (1.5.3).
• behaviour when executing unsupported <object> instances (2.1.2.1, 2.3.5).
• scope specified on individual form grammars takes precedence over the default grammar scope in a <form> (2.1).
• effect of <goto nextitem> on form item (2.1.5).
• no variables, conditions or counters are reset when using < goto nextitem> (2.1.5).
• behaviour of <transfer>, <subdialog> and <object> with audio playback in collect phase (2.1.6).
• enumerated executable context elements which cause execution to terminate (2.1.6.1).
• event handler selection in FIA process phase and <filled> (2.1.6.2).
• <reprompt> does not terminate the FIA (2.1.6.2.3).
• use of <enumerate> (2.2, 2.3.1)
• precedence of <choice> "next" attribute over "expr" attribute and precedence of "next" and "expr" over "event" attribute (2.2).
• specification of approximate grammar generation in <menu> and <choice> (2.2)
• <grammar> overrides automatically generated grammars in <choice> (2.2).
• <choice> expr evaluates to URI to transition to (2.2).
• <choice> event handler without control transition, causes menu to be re-executed (2.2).
• assignment of field variable when DTMF attribute defined (2.3.1).
• explanatory notes on portable use of builtins and expected platform dependence (2.3.1.1).
• field item variable names must respect ECMAScript variable naming conventions (2.3.1, 5.1).
• result value returned from "number" builtin type (2.3.1.1).
• Currency code not specified if not spoken (2.3.1.1)
• precedence of grammar "src" attribute over inline grammar (2.3.1.2)
• string returned when DTMF input when no "value" or CDATA specified in <option> (2.3.1.3).
• <option> and <grammar> can be used simultaneously to specify grammars in a <field> (2.3.1.3).
• use of builtin DTMF and speech grammars (2.3.1.4).
• handling of contradictory parameters to digits builtin (2.3.1.5).
• parameterization of builtin DTMF and speech grammars (2.3.1.5).
• scope of variables in <subdialog> (2.3.4).
• <subdialog> context is independent of its calling context (variable instances are not shared) but its context follows normal scoping rules for grammars, events, and variables (2.3.4).
• in <subdialog> use "expr" attribute to set variable if no corresponding <param> specified (2.3.4).
• user hangup during recording terminates recording normally; data recorded prior to hangup can be returned to server (2.3.6).
• interpretation of grammars in <record> (2.3.6).
• field variable in <record> is a reference to recorded audio. When submitting recorded data to a server, the "enctype" of <submit> should be set to "multipart/form-data" (2.3.6, 5.3.8).
• termination of <tranfer> by speech or DTMF returns near_end_disconnect status (2.3.7).
• value of "dest" attribute on <transfer> (2.3.7).
• the form item variable of <transfer> is undefined for blind transfer (2.3.7).
• <link>s may have no grammars (2.5).
• events throw by a <link> are handled by the best qualified <catch> in active scope (2.5).
• "xml:lang" attribute in <grammar> does not require multi-lingual support from platform (3.1)
• unsupported grammar language results in error.unsupported.language event being thrown (3.1.1)
• unsupported language lists can be indicated in the message variable of a <throw> (3.1.1).
• 'number' results as a string which ECMAScript will automatically convert to a number in a numerical expression; string must not use a leading zero (3.1.1)
• "src" attribute takes precedence over an inline grammar within the same element (3.1.1.3).
• implicit grammars (such as options) do not support weights (3.1.1.4).
• <grammar> "mode" attribute conflict with grammar mode (3.1.1.5).
• ongoing work on semantic attachments within <grammar> (3.1.5)
• enclosing <prompt> required if text contains speech synthesis markup (4.1.1).
• "xml:lang" attribute in <prompt> does not require multi-lingual support from platform (4.1.2)
• unsupported synthesis language results in error.unsupported.language event being thrown (4.1.2).
• 'alternate content' in <audio> (4.1.3).
• in <audio> "src" has precedence over "expr" attribute (4.1.3).
• <audio> element ignored if "expr" evaluates to null (4.1.3).
• Issue note that audio streaming is currently a platform optimization and a "stream" attribute may be added to <audio> to give developer explicit control over audio streaming (4.1.3).
• stand-alone <value> element is legal outside <prompt> (4.1.4).
• TTS expression on <value> may return SSML fragment (4.1.4).
• <value> can playback audio data when "expr" specifies a <record> field name (4.1.4).
• buffered DTMF input is deleted when "bargein" is false on <prompt> (4.1.5).
• relationship between prompt queueing and input collection (4.1.8).
• VoiceXML and ECMAScript variables are part of the same variable space;variables declared in ECMAScript can be used directly in VoiceXML (5.1).
• VoiceXML variable names, including field names, must follow ECMAScript naming rules; variable name declarations cannot contain a dot; the field name "a.b" is illegal (5.1).
• implicit ECMAScript variables are not accessible in VoiceXML (5.1.1).
• scope of variables (5.1.2).
• only some cond operators need to be escaped (5.1.3).
• evaluation of relative URLs against active document (5.2).
• VoiceXML does not generally specify when events are thrown (5.2.1).
• definition of "event" and "count" attributes of <catch> (5.2.2).
• no inherent limitations on <catch> in, for example, case of user hangup (5.2.2).
• <catch>'s event may be an empty string indicating that all events are to be caught (5.2.2).
• "as if by copy" catch inheritance (5.2, 5.2.4).
• catch element selection algorithm (5.2.4).
• defined prefix match as token match rather than string match (5.2.4).
• "error.badfetch" pre-defined error type (5.2.6)
• HTTPS is not the same protocol as HTTP (5.2.6)
• effect of <reprompt> in catch elements (5.3.6)
• behaviour of <reprompt> when in <catch> with final <goto> (5.3.6)
• effect of URI in <goto> on document variables (5.3.7)
• precedence defined between <goto> "next", "expr", "nextitem" and "expritem" attributes (5.3.7)
• variables declared in VoiceXML or ECMAScript can be submitted (5.3.8).
• <exit> does not throw an "exit" event (5.3.9)
• no "type" attribute of <script> (5.3.12).
• <script> evaluated along with <var> elements and field item variables in <form> (5.3.12)
• definition of "charset" in <script> (5.3.12)
• revised prefetch (6.1)
• effect of "fetchhint" attribute (6.1.1)
• caching policy selection (6.1.2)
• caching follows HTTP 1.1 cache correctness rules (6.1.2)
• format of platform-specific properties (6.3.1)
• definitions of "completetimeout" and "incompletetimeout" speech recognizer properties (6.3.2)
• definition of time designation values (6.5)

Miscellaneous

• Using tentative media types (e.g. "application/grammar+xml") submitted to IETF for approval
• language specifies set of required audio formats for <audio> and <record> (1.2.4).
• capabilities of conforming VoiceXML platform with respect to speech and dtmf grammars, audio, TTS, record, and transfer support (1.2.5).
• platforms should identify themselves with User-Agent HTTP header (1.2.5).
• description of how semantic interpretations are mapped to form variables (3.1.6).
• reserved the variable namespace "_$" for internal use (5.1).
• use of variables and form items with names "session", "application", "document", and "dialog" not recommended (5.1.2).
• revised definition of Conforming VoiceXML Processor, including requirement to support syntax and semantics of all elements described in this document (Appendix F).
• set of required audio formats (Appendix E).
• reusability appendix (Appendix K).

Appendix K — Reusability

K.1 Reusable dialog components

Definition: A packaged application fragment designed to be invoked by arbitrary applications or other Reusable Dialog Components. A Reusable Dialog Component (RDC) encapsulates the code for an interaction with the caller.

Reusable dialog components provide pre-packaged functionality "out-of-the-box" that enables developers to quickly build applications by providing standard default settings and behavior. They shield developers from having to worry about many of the intricacies associated with building a robust speech dialog, e.g., confidence score interpretation, error recovery mechanisms, prompting, etc. This behavior can be customized by a developer if necessary to provide application-specific prompts, vocabulary, retry settings, etc.

In this version of VoiceXML, the only authentic reusable component calling mechanisms are <subdialog> and <object>. Components called this way follow a model similar to subroutines in programming languages: the component is configured by a well-defined set of parameters passed to the component, the component has a relatively constrained interaction with the calling application, the component returns a well-defined result, and control returns automatically to the point from which the component was called. This has all the significant advantages of modularity, reentrancy, and easy reuse provided by subroutines. Of the two kinds of components, only <subdialog> components are guaranteed to be as portable as VoiceXML itself. On the other hand, <object> components may be able to package advanced, reusable functionality that has not yet been introduced into the standard.

K.2 Templates and samples

Although reusable dialog components have the advantages of modularity, reentrancy, and easy reuse as described above, the disadvantage of such components is that they must be designed very carefully with an eye to reuse, and even with the most careful of designs it is possible that the application developer will encounter situations for which the component cannot be easily configured to handle the application requirements. In addition, while the constrained interaction of a component with its calling environment makes it possible for the component designer to create a component that works predictably in disparate environments, it also may make the user's interaction with the component seem disconnected from the rest of the application.

In such situations the application developer may wish to reuse VoiceXML source code in the form of samples and templates - samples designed for easy customizability. Such code is more easily tailored for and integrated into a particular application, at the expense of modularity and rentrancy.

Such templates and samples can be created by separating interesting VoiceXML code from a main dialog and then distributing that code by copy for use in other dialogs. This form of reusability allows the user of the copied VoiceXML code to modify it as necessary and continue to use their modified version indefinitely.

VoiceXML facilitates this form of reusability by preserving the separation of state between form elements. In this regard, VoiceXML and HTML are similar. An HTML table can be copied from one HTML page to another because the table can be displayed regardless of the context before or after the table element.

Although parameterizability, modularity, and maintainability may be sacrificed with this approach, it has the advantage of being simple, quick, and eminently customizable.

Appendix L —Acknowledgements

This W3C specification is based upon VoiceXML 1.0 submitted by the VoiceXML Forum in May 2000. The VoiceXML Forum authors were: Linda Boyer, IBM; Peter Danielsen, Lucent Technologies; Jim Ferrans, Motorola; Gerald Karam, AT&T; David Ladd, Motorola; Bruce Lucas, IBM; Kenneth Rehor, Lucent Technologies.

This version was written with the participation of members of the W3C Voice Browser Working Group.The following have significantly contributed to writing this specification:

• Dan Burnett, Nuance Communications
• Deborah Dahl, Unisys
• Peter Danielsen, Lucent
• Martin Dragomirecky, Cisco
• Jim Ferrans, Motorola
• Andrew Hunt, SpeechWorks International
• Gerald Karam, AT&T
• Dave Ladd, Dynamicsoft
• Paul Lamere, Sun Microsystems
• Bruce Lucas, IBM
• Scott McGlashan, PipeBeach
• Mitsuru Oshima, General Magic
• Brad Porter, Tellme
• Ken Rehor, Nuance Communications
• Steph Tryphonas, Tellme

The Working Group would like to thank Dave Raggett and Jim Larson for their invaluable management support.