W3C work on Voice Interaction

Dave Raggett, W3C/Phone.com

Voice Portals Conference, London 11th October 2000

Email: dsr@w3.org

W3C Activity Lead for Voice Browsers, XHTML and XForms

---

World Wide Web Consortium

Leading the Web to its Full Potential

• Director – Tim Berners-Lee, inventor of the Web

• Formed in 1994, W3C is an international consortium of over 460 companies and institutions

• Develops specifications, guidelines, software, and tools, e.g.

  • HTML, CSS, XML, XSL, SMIL, SVG

• Current Staff of 60, budget $7M/year

• Working Groups, Interest Groups and Coordination Groups

W3C Voice Browser Working Group

Working Group formed in March 1999, following workshop in October 1998, and previous work in W3C’s Web Accessibility Initiative

Public working drafts on requirements for:

• Spoken dialogs
• Reusable dialog components
• Speech synthesis
• Speech grammars
• Natural language semantics
• Multi-modal dialogs

Voice Browser Working Group

Now working on drafting specifications, with public drafts available for:

• Speech Synthesis
• Speech Grammars

New drafts expected soon for:

• Speech Dialogs
• Statistical language models
• NL Semantics

Relationship to Other Groups

• VoiceXML Forum

  • VoiceXML Submission formed basis for W3C work
  • Education/Awareness activities
  • Conformance/Testing

• ETSI

  • Standards for distributed speech recognition

• ECTF (Enterprise Computer Telephony Forum)

  • Call control & Computer Telephony Services Platform

• DARPA Communicator

  • Research into advanced speech applications

Voice Browser Working Group

Voice Browser WG - Membership

Why Voice Interaction Is Valuable

• Access via any telephone

• Hands and eyes free operation

• Devices too small for displays and keyboards

• WAP-phones

• Palm-top organizers

• Universal messaging

Anatomy of a Voice Interface

• Application and user take turns to speak

• Form filling metaphor

• Prompt user for each field in turn using synthetic speech and prerecorded audio

• Use speech grammars to interpret what user says

• Offer help as needed

• Submit completed form to back-end server

• Links to other “pages”

• Break out to scripting as needed

Speech Synthesis

• When to use prerecorded human speech

• Speech Synthesis engines are smart

• Basic properties: volume, rate, pitch

• Speech font selection by name, gender, age

• Control over how things are pronounced

• Prerecorded audio effects

• W3C Speech synthesis markup language

Speech Grammars

Context free grammars describe what user says, each rule associated with a semantic effect

[I want to fly to] $City { destination = $City }
$City = London | Paris | Amsterdam | Milan

Voice Dialog Example

This example dialog proceeds sequentially:

C (computer): Welcome to the international weather service.
  What country?

H (human): Help

C: Please say the country for which you want the weather.

H: France

C: What city?

H: Antibes

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

H: Cannes

C: The conditions in Cannes France are sunny and clear at 11 AM … 

Voice Dialog Markup

<form id="weather_info">       
  <block>Welcome to the international weather service.</block>       
  <field name=“country">       
    <prompt>What country?</prompt>       
    <grammar src=“country.gram" type="application/x-jsgf"/>       
    <catch event="help">       
      Please say the country for which you want the weather.       
    </catch>       
  </field>       
  <field name="city">       
    <prompt>What city?</prompt>       
    <grammar src="city.gram" type="application/x-jsgf"/>       
    <catch event="help">       
      Please say the city for which you want the weather.       
    </catch>       
  </field>       
  <block>       
    <submit next="/servlet/weather" namelist="city country"/>       
  </block>       
</form> 

Keys to Successful Voice Apps

• Keep it simple!

• Avoid deeply nested task contexts

• Carefully consider when a human touch is needed – and handover to human operator

• There’s no substitute for user testing

• Plan for continuous improvement based on user feedback

Where Next?

• Improvements to VoiceXML

• Multi-modal applications

  • Voice + Display + Key pad etc.

  • User is free to switch between voice interaction and use of display/key pad

  • Imagine combination of WML+VoiceXML

• W3C plans to set up Multi-modal Dialog Working Group in 2001

Copyright © 1994-2000 W3C^® (MIT, INRIA, Keio), All Rights Reserved. W3C liability, trademark, document use and software licensing rules apply. Your interactions with this site are in accordance with our public and Member privacy statements.