W3C

CSS3 Speech Module

W3C Working Draft 19 April 2011

This version:
http://www.w3.org/TR/2011/WD-css3-speech-20110419
Latest version:
http://www.w3.org/TR/css3-speech
Previous versions:
http://www.w3.org/TR/2004/WD-css3-speech-20041216/
Editors:
Dave Raggett (W3C/Canon)
Daniel Glazman (Disruptive Innovations)
Claudio Santambrogio (Opera Software)
Daniel Weck (DAISY Consortium)

Abstract

CSS (Cascading Style Sheets) is a language for describing the rendering of HTML and XML documents on screen, on paper, in speech, etc. CSS defines aural properties that give control over rendering XML and HTML to speech. This draft describes the text to speech properties proposed for CSS level 3, and is a re-work of the informative CSS2.1 Aural appendix [CSS21]. These are designed for match the model described in the Speech Synthesis Markup Language (SSML) Version 1.0 [SPEECH-SYNTHESIS].

The CSS3 Speech Module is a community effort and if you would like to help with implementation and driving the specification forward along the W3C Recommendation track, please contact the editors.

Status of this document

This section describes the status of this document at the time of its publication. Other documents may supersede this document. A list of current W3C publications and the latest revision of this technical report can be found in the W3C technical reports index at http://www.w3.org/TR/.

Publication as a Working Draft does not imply endorsement by the W3C Membership. This is a draft document and may be updated, replaced or obsoleted by other documents at any time. It is inappropriate to cite this document as other than work in progress.

The (archived) public mailing list www-style@w3.org (see instructions) is preferred for discussion of this specification. When sending e-mail, please put the text “css3-speech” in the subject, preferably like this: “[css3-speech] …summary of comment…

This document was produced by the CSS Working Group (part of the Style Activity).

This document was produced by a group operating under the 5 February 2004 W3C Patent Policy. W3C maintains a public list of any patent disclosures made in connection with the deliverables of the group; that page also includes instructions for disclosing a patent. An individual who has actual knowledge of a patent which the individual believes contains Essential Claim(s) must disclose the information in accordance with section 6 of the W3C Patent Policy.


This document is a draft of one of the "modules" for the upcoming CSS3 specification.

This document has been developed in cooperation with the Voice Browser working group (W3C Members only).

 

The following issues need to be discussed and require working group resolutions:

The CSS WG maintains an issues list for this module.

Table of contents


1. Dependencies on other modules

This CSS3 module depends on the following other CSS3 modules:

It has non-normative (informative) references to the following other CSS3 modules:

2. Introduction

The speech rendering of a document, already commonly used by the blind and print-impaired communities, combines speech synthesis and "auditory icons". Often such aural presentation occurs by converting the document to plain text and feeding this to a screen reader — software or hardware that simply reads all the characters on the screen. This results in less effective presentation than would be the case if the document structure were retained. Style sheet properties for text to speech may be used together with visual properties (mixed media) or as an aural alternative to visual presentation.

Besides the obvious accessibility advantages, there are other large markets for listening to information, including in-car use, industrial and medical documentation systems (intranets), home entertainment, and to help users learning to read or who have difficulty reading.

When using voice properties, the canvas consists of a two channel stereo space and a temporal space (you can specify audio cues before and after synthetic speech). The CSS properties also allow authors to vary the characteristics of synthetic speech (voice type, frequency, inflection, etc.).

Examples:

h1, h2, h3, h4, h5, h6 {
    voice-family: paul;
    voice-stress: moderate;
    cue-before: url(ping.au)
}
p.heidi { voice-balance: left; voice-family: female }
p.peter { voice-balance: right; voice-family: male }
p.goat  { voice-volume: soft }

This will direct the speech synthesizer to speak headers in a voice (a kind of "audio font") called "paul". Before speaking the headers, a sound sample will be played from the given URL. Paragraphs with class "heidi" will appear to come from the left (if the sound system is capable of stereo), and paragraphs of class "peter" from the right. Paragraphs with class "goat" will be played softly.

Note that content creators may conditionally include CSS properties authored specifically for user-agents with text to speech capabilities (TTS), by specifying the "speech" media type via the media attribute of the link element, or with the @media at-rule, or within an @import statement (the "aural" media type was deprecated in the informative CSS2.1 Aural appendix [CSS21]). When doing so, the styles authored within the scope of such conditional statements are ignored by user-agents that do not support speech synthesis.

3. The aural "box" model

The formatting model of CSS for aural media is based on a sequence of sounds and silences that appear in a nested model which is related to the visual box model; however the aural canvas is one-dimensional, monolinear. For compatibility with the visual box model, we will call it the aural "box" model.

The element is surrounded by, in this order, ‘rest’, ‘cue’ and ‘pause’ properties - they can be thought of as aural equivalents to ‘padding’, ‘border’ and ‘margin’ respectively.

It can be represented in the following way (including the equivalent properties from the visual box model for clarification of relationships):

A graph depicting the aural 'box' model.

where <element> is the selected element to which the properties from the CSS3 Speech Module apply.

Note that the ‘none’ value of the ‘display’ property (which is defined in the CSS box model [CSS3BOX] and which applies to all media types) influences the resolved value of the ‘speakability’ property (defined within this CSS3 module) when the ‘auto’ value is specified for ‘speakability’. This is the only case whereby a property defined externally to this CSS3 module affects a characteristic specific to the aural dimension (i.e. has an impact within the aural "box" model).

4. Mixing properties: ‘voice-volume’ and ‘voice-balance

Name: voice-volume
Value: <non-negative number> | <percentage> | silent | x-soft | soft | medium | loud | x-loud | inherit
Initial: medium
Applies to: all elements
Inherited: yes
Percentages: refer to inherited value
Media: speech

The ‘voice-volume’ refers to the amplitude of the waveform output by the speech synthesiser. This may be mixed with other audio sources, influencing the perceived loudness of synthetic speech relative to these sources.

Values have the following meanings:

<non-negative number>
An integer or floating point number in the range ‘0’ to ‘100’. ‘0’ represents silence (the minimum level), and 100 corresponds to the maximum level. The volume scale is linear amplitude.
<percentage>
Only positive values are allowed. Computed values are calculated relative to the inherited value, and are then clipped to the range ‘0’ to ‘100’.

Note that a leading "+" sign does not denote an increment. For example, +50% is equivalent to 50%, so the computed value equals the inherited value times 0.5 (divided by 2), then clipped to [0,100].

silent, x-soft, soft, medium, loud, and x-loud
A sequence of monotonically non-decreasing volume levels. The value of ‘silent’ is mapped to ‘0’ and ‘x-loud’ is mapped to ‘100’. The mapping of other values to numerical volume levels is implementation dependent and may vary from one speech synthesizer to the next.

User agents should allow the level corresponding to ‘100’ to be set by the listener. No one setting is universally applicable; suitable values depend on the equipment in use (speakers, headphones), and the environment (in car, home theater, library) and personal preferences.

Note that there is a difference between an element whose ‘voice-volume’ property has a value of ‘silent’, and an element whose ‘speakability’ property has the value ‘none’. The former takes up the same time as if it had been spoken, including any pause before and after the element, but no sound is generated (although descendants may override the ‘voice-volume’ value and may therefore generate audio output). The latter requires no time and is not rendered in the aural dimension (although descendants may override the ‘speakability’ value and may therefore generate audio output).

Name: voice-balance
Value: <number> | left | center | right | leftwards | rightwards | inherit
Initial: center
Applies to: all elements
Inherited: yes
Percentages: N/A
Media: speech

The ‘voice-balance’ property refers to the balance between left and right channels, and presumes a two channel (stereo) model that is widely supported on consumer audio equipment.

Values have the following meanings:

<number>
An integer or floating point number between ‘-100’ and ‘100’. For ‘-100’ only the left channel is audible. Simarly for ‘100’ or ‘+100’ only the right channel is audible. For ‘0’ both channels have the same level, so that the speech appears to be coming from the center.
left
Same as ‘-100’.
center
Same as ‘0’.
right
Same as ‘100’ or ‘+100’.
leftwards
Moves the sound to the left, relative to the inherited ‘voice-balance’. More precisely, subtract 20 from the inherited value and clip the resulting value to the range ‘-100’ and ‘100’.
rightwards
Moves the sound to the right, relative to the inherited ‘voice-balance’. More precisely, add 20 to the inherited value and clip the resulting value to the range ‘-100’ and ‘100’.

Many speech synthesizers only support a single channel. The ‘voice-balance’ property can then be treated as part of a post synthesis mixing step. This is where speech is mixed with other audio sources.

5. Speaking properties: ‘speakability’, ‘speak

Name: speakability
Value: auto | none | normal | inherit
Initial: auto
Applies to: all elements
Inherited: yes
Percentages: N/A
Media: speech

ISSUE: should the "speakability" property be named differently ? e.g. "speaking"

This property specifies whether text will be rendered aurally.

Values have the following meanings:

auto
Resolves to a computed value of ‘none’ when ‘display’ is ‘none’ (see [CSS3BOX]), otherwise resolves to a computed value of ‘auto’ which yields a used value of ‘normal’.
none
This value causes an element (including pauses, cues, rests and actual content) to not be rendered (i.e., the element has no effect in the aural dimension).

Note that any of the descendants of the affected element are allowed to override this value, so they may actually take part in the aural rendering. However, the pauses, cues, and rests of the ancestor element remain "deactivated" in the aural dimension, and therefore do not contribute to the collapsing of pauses or additive behavior of adjacent rests.

normal
The element is rendered aurally.

Note that although the ‘none’ value of the ‘display’ property cannot be overridden by descendants of the affected element (see [CSS3BOX]), the ‘auto’ value of ‘speakability’ can however be overridden by descendants, using either of ‘none’ or ‘normal’. In the case of ‘normal’, this would result in descendants being rendered in the aural dimension even though they would not be rendered on the visual canvas.

Name: speak
Value: normal | spell-out | digits | literal-punctuation | no-punctuation | inherit
Initial: normal
Applies to: all elements
Inherited: yes
Percentages: N/A
Media: speech

This property specifies in what manner text gets rendered aurally.

Values have the following meanings:

normal
Uses language-dependent pronunciation rules for rendering an element and its children. Punctuation is not to be spoken, but instead rendered naturally as various pauses.
spell-out
Spells the text one letter at a time (useful for acronyms and abbreviations). In languages where accented characters are rare, it is permitted to drop accents in favor of alternative unaccented spellings. As as example, in English, the word "rôle" can also be written as "role". A conforming implementation would thus be able to spell-out "rôle" as "R O L E".
digits
Speak numbers one digit at a time, for instance, "twelve" would be spoken as "one two", and "31" as "three one".
literal-punctuation
Similar to ‘normal’ value, but punctuation such as semicolons, braces, and so on are to be spoken literally.
no-punctuation
Similar to ‘normal’ value but punctuation is not to be spoken nor rendered as various pauses.

Speech synthesizers are knowledgeable about what is and what is not a number. The ‘speak’ property gives authors the means to control how the synthesizer renders the numbers it discovers in the source text, and may be implemented as a preprocessing step before passing the text to the speech synthesizer.

6. Pause properties: ‘pause-before’, ‘pause-after’ and ‘pause

Name: pause-before
Value: <time> | none | x-weak | weak | medium | strong | x-strong | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

 

Name: pause-after
Value: <time> | none | x-weak | weak | medium | strong | x-strong | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

These properties specify a pause or prosodic boundary to be observed before (or after) an element or, if any ‘cue-before’ (or ‘cue-after’) is specified, before (or after) these.

Values have the following meanings:

<time>
Expresses the pause in absolute time units, as per the syntax of "time" values in [CSS3VAL] (seconds and milliseconds, e.g. "3s", "250ms"). Only positive values are allowed.
none, x-weak, weak, medium, strong, and x-strong
These values may be used to indicate the prosodic strength of the break in speech output. The synthesis processor may insert a pause as part of its implementation of the prosodic break. The value "none" indicates that no prosodic break boundary should be output, and can be used to inhibit a prosodic break which the processor would otherwise produce. The other values indicate monotonically non-decreasing (conceptually increasing) break strength between elements. "x-weak" and "x-strong" are mnemonics for "extra weak" and "extra strong", respectively. The stronger boundaries are typically accompanied by pauses. The breaks between paragraphs are typically much stronger than the breaks between words within a sentence.
Name: pause
Value: [ <‘pause-before’> || <‘pause-after’> ] | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

The ‘pause’ property is a shorthand for setting ‘pause-before’ and ‘pause-after’. If two values are given, the first value is ‘pause-before’ and the second is ‘pause-after’. If only one value is given, it applies to both properties.

Examples:

h1 { pause: 20ms } /* pause-before: 20ms; pause-after: 20ms */
h2 { pause: 30ms 40ms } /* pause-before: 30ms; pause-after: 40ms */
h3 { pause-after: 10ms } /* pause-before: unspecified; pause-after: 10ms */

6.1. Collapsing pauses

The pause defines the minimum distance of the aural "box" to the aural "boxes" before and after it. Adjacent pauses should be merged by selecting the strongest named break or the longest absolute time interval. Thus "strong" is selected when comparing "strong" and "weak", while "1s" is selected when comparing "1s" and "250ms". We say that such pauses collapse. A combination of a named break and time duration is treated additively.

The following pauses collapse:

  1. The ‘pause-after’ of an aural "box" and the ‘pause-after’ of its last child, provided the former has no ‘rest-after’ and no ‘cue-after’.
  2. The ‘pause-before’ of an aural "box" and the ‘pause-before’ of its first child, provided the former has no ‘rest-before’ and no ‘cue-before’.
  3. The ‘pause-after’ of an aural "box" and the ‘pause-before’ of its next sibling.
  4. The ‘pause-before’ and ‘pause-after’ of an aural "box", if the the "box" has a ‘voice-duration’ of "0ms", no ‘rest-before’ or ‘rest-after’, no ‘cue-before’ or ‘cue-after’ and no content.

The ‘pause-after’ of an element is always adjoining to the ‘pause-before’ of its next sibling.

The ‘pause-before’ an element is adjoining to its first child's ‘pause-before’, if the element has no ‘cue-before’ nor ‘rest-before’.

The ‘pause-after’ of an element is adjoining to its last child's ‘pause-after’, if the element has no ‘rest-after’ or ‘cue-after’.

An element's own pauses are adjoining if the ‘voice-duration’ property is zero, and it has no rest, and it does not contain any content, and all of its children's pauses (if any) are adjoining.

Note that ‘pause’ has been moved from between the element's contents and any ‘cue’ to outside the ‘cue’. This is not backwards compatible with the informative CSS2.1 Aural appendix [CSS21].

7. Rest properties: ‘rest-before’, ‘rest-after’ and ‘rest

Name: rest-before
Value: <time> | none | x-weak | weak | medium | strong | x-strong | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

 

Name: rest-after
Value: <time> | none | x-weak | weak | medium | strong | x-strong | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

These properties specify a rest or prosodic boundary to be observed before (or after) speaking an element's content.

Values have the following meanings:

<time>
Expresses the rest in absolute time units, as per the syntax of "time" values in [CSS3VAL] (seconds and milliseconds, e.g. "3s", "250ms"). Only positive values are allowed.
none, x-weak, weak, medium, strong, and x-strong
These values may be used to indicate the prosodic strength of the break in speech output. The synthesis processor may insert a rest as part of its implementation of the prosodic break. The value "none" indicates that no prosodic break boundary should be output, and can be used to inhibit a prosodic break which the processor would otherwise produce. The other values indicate monotonically non-decreasing (conceptually increasing) break strength between words. The stronger boundaries are typically accompanied by rests. "x-weak" and "x-strong" are mnemonics for "extra weak" and "extra strong", respectively.

As opposed to pause properties, the rest is inserted between the element's content and any ‘cue-before’ or ‘cue-after’ content. Adjacent rests are treated additively.

Name: rest
Value: [ <‘rest-before’> || <‘rest-after’> ] | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

The ‘rest’ property is a shorthand for setting ‘rest-before’ and ‘rest-after’. If two values are given, the first value is ‘rest-before’ and the second is ‘rest-after’. If only one value is given, it applies to both properties.

8. Cue properties: ‘cue-before’, ‘cue-after’ and ‘cue

Name: cue-before
Value: <uri> [<percentage>] | none | inherit
Initial: none
Applies to: all elements
Inherited: no
Percentages: apply to inherited value for ‘voice-volume
Media: speech

 

Name: cue-after
Value: <uri> [<percentage>] | none | inherit
Initial: none
Applies to: all elements
Inherited: no
Percentages: apply to inherited value for ‘voice-volume
Media: speech

Auditory icons are another way to distinguish semantic elements. Sounds may be played before and/or after the element to delimit it.

Values have the following meanings:

<uri>
The URI must designate an auditory icon resource. If the URI resolves to something other than an audio file, such as an image, the resource should be ignored and the property treated as if it had the value ‘none’.
none
No auditory icon is specified.
<percentage>
Only positive values are allowed. Computed values are calculated relative to the inherited value of the ‘voice-volume’ property, and are then clipped to the range ‘0’ to ‘100’. ‘0’ represents silence (the minimum level), and 100 corresponds to the maximum level. The volume scale is linear amplitude. By basing the percentage upon the inherited value for ‘voice-volume’, it is easy to adjust the relative loudness of cues compared to synthetic speech for whatever volume setting has been provided for that speech.

Note that a leading "+" sign does not denote an increment. For example, +50% is equivalent to 50%, so the computed value equals the inherited value times 0.5 (divided by 2), then clipped to [0,100].

Examples:

a { cue-before: url(bell.aiff); cue-after: url(dong.wav) }

h1 { cue-before: url(pop.au) 80%; cue-after: url(pop.au) 50% }

div.caution { cue-before: url(caution.wav) 130% }
Name: cue
Value: [ <‘cue-before’> || <‘cue-after’> ] | inherit
Initial: not defined for shorthand properties
Applies to: all elements
Inherited: no
Percentages: apply to inherited value for ‘voice-volume
Media: speech

The ‘cue’ property is a shorthand for setting ‘cue-before’ and ‘cue-after’. If two values are given the first value is ‘cue-before’ and the second is ‘cue-after’. If only one value is given, it applies to both properties.

The following two rules are equivalent:

h1 {cue-before: url(pop.au); cue-after: url(pop.au) }

h1 {cue: url(pop.au) }

If a user agent cannot render an auditory icon (e.g., the user's environment does not permit it), we recommend that it produce an alternative audio cue (e.g., popping up a warning, emitting a warning sound, etc.)

Authors may also use content generation techniques to insert additional auditory cues based on text instead of audio icons.

9. Voice characteristic properties: ‘voice-family’, ‘voice-rate’, ‘voice-pitch’, ‘voice-pitch-range’ and ‘voice-stress

Name: voice-family
Value: [[[<specific-voice> | [<age>? <generic-voice>]] <non-negative number>?],]* [[<specific-voice> | [<age>? <generic-voice>]] <non-negative number>?] | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: yes
Percentages: N/A
Media: speech

The value is a comma-separated, prioritized list of voice family names (compare with font-family).

Values have the following meanings:

<specific-voice>
Values are specific instances (e.g., Mike, comedian, mary, carlos, "valley-girl").
<age>
Possible values are ‘child’, ‘young’ and ‘old’.
<generic-voice>
Values are voice families. Possible values are ‘male’, ‘female’ and ‘neutral’.
<non-negative number>
Indicates a preferred variant of the other voice characteristics (e.g. "the second male child voice named ‘Mike’"). Possible values are positive integers, excluding zero (i.e. starting from 1). The value "1" refers to the first of all matching voices.

Examples:

h1 { voice-family: announcer, old male }
p.part.romeo { voice-family: romeo, young male }
p.part.juliet { voice-family: juliet, female }
p.part.mercutio { voice-family: male 2 }
p.part.tybalt { voice-family: male 3 }
p.part.nurse { voice-family: child female }

Names of specific voices may be quoted, and indeed must be quoted if any of the words that make up the name does not conform to the syntax rules for identifiers [CSS3SYN]. Any whitespace characters before and after the voice name are ignored. For compatibility with SSML, whitespace characters are not permitted within voice names.

The ‘voice-family’ property is used to guide the selection of the voice to be used for speech synthesis. The overriding priority is to match the language specified by the xml:lang attribute as per the XML 1.0 specification [XML10], and as inherited by nested elements until overridden by a further xml:lang attribute.

If there is no voice available for the requested value of xml:lang, the processor should select a voice that is closest to the requested language (e.g. a variant or dialect of the same language). If there are multiple such voices available, the processor should use a voice that best matches the values provided with the ‘voice-volume’ property. It is an error if there are no such matches.

Name: voice-rate
Value: <percentage> | x-slow | slow | medium | fast | x-fast | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: yes
Percentages: refer to default value
Media: speech

This property controls the speaking rate. The default rate for a voice depends on the language and dialect and on the personality of the voice. The default rate for a voice should be such that it is experienced as a normal speaking rate for the voice when reading aloud text. Since voices are processor-specific, the default rate will be as well.

Values have the following meanings:

<percentage>
Only positive values are allowed. Computed values are calculated relative to the default speaking rate for each voice.

Note that a leading "+" sign does not denote an increment, for example +50% is equivalent to 50% (i.e. the computed value equals the inherited value times 0.5, which is half the normal rate of the voice).

x-slow, slow, medium, fast and x-fast
A sequence of monotonically non-decreasing speaking rates that are implementation and voice specific.
Name: voice-pitch
Value: <frequency> | <percentage> | <relative-change> | x-low | low | medium | high | x-high | inherit
Initial: medium
Applies to: all elements
Inherited: yes
Percentages: refer to inherited value
Media: speech

Specifies the average pitch (a frequency) of the speaking voice. The average pitch of a voice depends on the ‘voice-family’. For example, the average pitch for a standard male voice is around 120Hz, but for a female voice, it's around 210Hz.

Values have the following meanings:

<frequency>
This is an absolute value that specifies the average pitch of the speaking voice in Hertz. It must be <non-negative number> followed by the "Hz" suffix.
<percentage>
Only positive values are allowed. Computed values are calculated relative to the inherited value.

Note that a leading "+" sign does not denote an increment. For example, +50% is equivalent to 50%, so the computed value equals the inherited value times 0.5 (divided by 2), which is half the inherited average pitch of the voice.

<relative-change>
Specifies a relative change (decrement or increment) to the inherited value. The syntax of allowed values is a <number> (the "+" sign is optional for positive numbers), followed by either of "Hz" (for Hertz) or "st" (for semitones), and followed by a space character and the "relative" keyword.

Note that the "relative" keyword is mandatory. This is in order to disambiguate from <frequency> values which may also carry the optional "+" sign on positive values.

x-low, low, medium, high, x-high
A sequence of monotonically non-decreasing pitch levels that are implementation and voice specific.

Examples:

h1 { voice-pitch: 250Hz; }
h1 { voice-pitch: +250Hz; } /* identical to the line above */
h2 { voice-pitch: +30Hz relative; }
h2 { voice-pitch: 30Hz relative; } /* identical to the line above */
h3 { voice-pitch: -2st relative; }
h4 { voice-pitch: -2st; } /* Illegal syntax ! ("relative" keyword is missing) */
Name: voice-pitch-range
Value: <frequency> | <percentage> | <relative-change> | x-low | low | medium | high | x-high | inherit
Initial: implementation dependent
Applies to: all elements
Inherited: yes
Percentages: refer to inherited value
Media: speech

Specifies variation in average pitch. The perceived pitch of a human voice is determined by the fundamental frequency and typically has a value of 120Hz for a male voice and 210Hz for a female voice. Human languages are spoken with varying inflection and pitch; these variations convey additional meaning and emphasis. Thus, a highly animated voice, i.e., one that is heavily inflected, displays a high pitch range. This property specifies the range over which these variations occur, i.e., how much the fundamental frequency may deviate from the average pitch.

Values have the following meanings:

<frequency>
This is an absolute value that specifies the average pitch range of the speaking voice in Hertz. Low ranges produce a flat, monotonic voice. A high range produces animated voices. It must be <non-negative number> followed by the "Hz" suffix.
<percentage>
Only positive values are allowed. Computed values are calculated relative to the inherited value.

Note that a leading "+" sign does not denote an increment. For example, +50% is equivalent to 50%, so the computed value equals the inherited value times 0.5 (divided by 2), which is half the inherited average pitch range of the voice.

<relative-change>
Specifies a relative change (decrement or increment) to the inherited value. The syntax of allowed values is a <number> (the "+" sign is optional for positive numbers), followed by either of "Hz" (for Hertz) or "st" (for semitones), and followed by a space character and the "relative" keyword.

Note that the "relative" keyword is mandatory. This is in order to disambiguate from <frequency> values which may also carry the optional "+" sign on positive values.

x-low, low, medium, high and x-high
A sequence of monotonically non-decreasing pitch ranges that are implementation and language dependent.

Note that a semitone is half of a tone (a half step) on the standard diatonic scale. A semitone doesn't correspond to a fixed value in Hertz: instead, the ratio between two consecutive frequencies separated by exactly one semitone is approximately 1.05946 (the actual arithmetics involved are beyond the scope of this specification, please refer to existing literature on that subject).

Name: voice-stress
Value: strong | moderate | none | reduced | inherit
Initial: moderate
Applies to: all elements
Inherited: yes
Percentages: N/A
Media: speech

Indicates the strength of emphasis to be applied. Emphasis is indicated using a combination of pitch change, timing changes, loudness and other acoustic differences) that varies from one language to the next.

Values have the following meanings:

none, moderate and strong
These are monotonically non-decreasing in strength, with the precise meanings dependent on language being spoken. The value ‘none’ inhibits the synthesizer from emphasizing words it would normally emphasize.
reduced
Effectively the opposite of emphasizing a word. For example, when the phrase "going to" is reduced it may be spoken as "gonna".

10. Duration property: ‘voice-duration

Name: voice-duration
Value: <time>
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

Allows authors to specify how long it should take to render the selected element's content. This property takes precedence over the ‘voice-rate’ property.

Values have the following meanings:

<time>

Specifies a value in seconds or milliseconds (as per the syntax of "time" values in [CSS3VAL]) for the desired spoken duration of the element contents, for instance, "250ms", or "3s". Only positive numbers are allowed.

11. Phonetics: ‘phonemes’ and @phonetic-alphabet

Name: phonemes
Value: <string>
Initial: implementation dependent
Applies to: all elements
Inherited: no
Percentages: N/A
Media: speech

This allows authors to specify a phonetic pronunciation for the text contained by the corresponding element.

Values have the following meanings:

<string>
A character string that describes a pronunciation based on the specified @phonetic-alphabet.

ISSUE: the ‘phonemes’ property covers functionality that arguably doesn't belong to CSS, as it breaks the principle of separation between content and presentation (i.e. any change in the source text must be echoed by an corresponding change in the content of the aural ‘phonemes’ property). The broader aspect of pronunciation lexicons would be better addressed in the content markup. For example, it is proposed that formats like the W3C's own PLS standard should be supported by the link/rel mechanism. There is a related discussion on the public mailing-list.

11.1. @phonetic-alphabet

A phonetic alphabet is a collection of symbols that represent the sounds of one or more human languages. The default alphabet for the pronunciation string of the ‘phonemes’ property is the International Phonetic Alphabet ("ipa"), corresponding to Unicode representations of the phonetic characters developed by the International Phonetic Association [IPA]. The phonetic alphabet can be explicitly specified using the @phonetic-alphabet rule.

Note that the alphabet is specified via an at-rule to avoid problems with inappropriate cascades [CSS3CASCADE] that can occur if the alphabet was set via a property.

Example:

@phonetic-alphabet "ipa";
#tomato { phonemes: "t\0252 m\0251 to\028a " }

This will direct the speech synthesizer to replace the default pronunciation by the corresponding sequence of ‘phonemes’ in the designated alphabet.

At most one @phonetic-alphabet rule may appear in style sheet and it must appear, when used, before any occurrence of the ‘phonemes’ property. The only valid values are the default "ipa" phonetic alphabet, and vendor-specific strings such as "x-organization" or "x-organization-alphabet".

12. Inserted and replaced content

Sometimes, authors will want to specify a mapping from the source text into another string prior to the application of the regular pronunciation rules. This may be used for uncommon abbreviations or acronyms which are unlikely to be recognized by the synthesizer. The ‘content’ property can be used to replace one string by another. In the following example, the abbreviation is rendered using the content of the title attribute instead of the element's content:

Example:

abbr { content: attr(title); }
  ...

<abbr title="World Wide Web Consortium">W3C</abbr>

This replaces the content of the selected element by the string "World Wide Web Consortium".

In a similar way text-to-speech strings in a document can be replaced by a previously recorded version:

Example:

.hamlet { content: url(gielgud.wav); }
  ...

<div class="hamlet">
To be, or not to be: that is the question:
</div>

If the format is supported, the file is available and the UA is configured to do so, a recording of Sir John Gielgud's declamation of the famous monologue will be played, otherwise the UA falls back to render the text-to-speech with its own synthesizer.

Furthermore, authors (or users in a user stylesheet) may want to add some information to ease understanding the structure for non-visual interaction with the document. They can do so by using the ‘::before’ and ‘::after’ pseudo-elements that will be inserted between the element's contents and the ‘rest’:

Example:

ul::before { content: "Start list: " }
ul::after  { content: "List end. " }
li::before { content: "List item: " }

This inserts the string "Start list: " before a list and the string "List item: " before each list item; likewise the string "List end: " inserted after the list will inform the user that the list is finished.

Different stylesheets can be used to define the level of verbosity for additional information spoken by screen readers.

Note that detailed information can be found in the CSS3 Generated and Replaced Content Module [CSS3GENCON].

ISSUE: the speech handling of list items is under-specified. One suggestion is to ignore list-style-type (which can represent visual glyphs not suitable for aural processing anyway) and to define a standard way to read list "markers". Unfortunately this introduces a dependency on a feature specific to CSS3-Lists. There is a related discussion on the public mailing-list.


Appendix A — Profiles

TBD

Appendix B — Property index

Property Values Initial Applies to Inh. Percentages Media
cue [ <‘cue-before’> || <‘cue-after’> ] | inherit not defined for shorthand properties all elements no apply to inherited value for ‘voice-volume’ speech
cue-after <uri> [<percentage>] | none | inherit none all elements no apply to inherited value for ‘voice-volume’ speech
cue-before <uri> [<percentage>] | none | inherit none all elements no apply to inherited value for ‘voice-volume’ speech
pause [ <‘pause-before’> || <‘pause-after’> ] | inherit implementation dependent all elements no N/A speech
pause-after <time> | none | x-weak | weak | medium | strong | x-strong | inherit implementation dependent all elements no N/A speech
pause-before <time> | none | x-weak | weak | medium | strong | x-strong | inherit implementation dependent all elements no N/A speech
phonemes <string> implementation dependent all elements no N/A speech
rest [ <‘rest-before’> || <‘rest-after’> ] | inherit implementation dependent all elements no N/A speech
rest-after <time> | none | x-weak | weak | medium | strong | x-strong | inherit implementation dependent all elements no N/A speech
rest-before <time> | none | x-weak | weak | medium | strong | x-strong | inherit implementation dependent all elements no N/A speech
speak normal | spell-out | digits | literal-punctuation | no-punctuation | inherit normal all elements yes N/A speech
speakability auto | none | normal | inherit auto all elements yes N/A speech
voice-balance <number> | left | center | right | leftwards | rightwards | inherit center all elements yes N/A speech
voice-duration <time> implementation dependent all elements no N/A speech
voice-family [[[<specific-voice> | [<age>? <generic-voice>]] <non-negative number>?],]* [[<specific-voice> | [<age>? <generic-voice>]] <non-negative number>?] | inherit implementation dependent all elements yes N/A speech
voice-pitch <frequency> | <percentage> | <relative-change> | x-low | low | medium | high | x-high | inherit medium all elements yes refer to inherited value speech
voice-pitch-range <frequency> | <percentage> | <relative-change> | x-low | low | medium | high | x-high | inherit implementation dependent all elements yes refer to inherited value speech
voice-rate <percentage> | x-slow | slow | medium | fast | x-fast | inherit implementation dependent all elements yes refer to default value speech
voice-stress strong | moderate | none | reduced | inherit moderate all elements yes N/A speech
voice-volume <non-negative number> | <percentage> | silent | x-soft | soft | medium | loud | x-loud | inherit medium all elements yes refer to inherited value speech

The following properties are defined in other modules:

Appendix C — Index

Appendix D — Definitions

Glossary

The following terms and abbreviations are used in this module.

UA
User Agent

A program that reads and/or writes CSS style sheets on behalf of a user in either or both of these categories: programs whose purpose is to render documents (e.g., browsers) and programs whose purpose is to create style sheets (e.g., editors). A UA may fall into both categories. (There are other programs that read or write style sheets, but this module gives no rules for them.)

document

A tree-structured document with elements and attributes, such as an SGML or XML document [XML11].

style sheet

A CSS style sheet.

Conformance

Conformance requirements are expressed with a combination of descriptive assertions and RFC 2119 terminology. The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in the normative parts of this document are to be interpreted as described in RFC 2119. However, for readability, these words do not appear in all uppercase letters in this specification. All of the text of this specification is normative except sections explicitly marked as non-normative, examples, and notes. [RFC2119]

Examples in this specification are introduced with the words "for example" or are set apart from the normative text with class="example", like this:

This is an example of an informative example.

Informative notes begin with the word "Note" and are set apart from the normative text with class="note", like this:

Note, this is an informative note.

Conformance to the CSS3 Speech Module is defined for three classes:

style sheet
A CSS style sheet.
renderer
A UA that interprets the semantics of a style sheet and renders documents that use them.
authoring tool
A UA that writes a style sheet.

A style sheet is conformant to the CSS3 Speech Module if all of its declarations that use properties defined in this module have values that are valid according to the generic CSS grammar and the individual grammars of each property as given in this module.

A renderer is conformant to the CSS3 Speech Module if, in addition to interpreting the style sheet as defined by the appropriate specifications, it supports all the properties defined by CSS3 Speech Module by parsing them correctly and rendering the document accordingly. However the inability of a UA to correctly render a document due to limitations of the device does not make the UA non-conformant. (For example, a UA is not required to render color on a monochrome monitor.)

An authoring tool is conformant to CSS3 Speech Module if it writes syntactically correct style sheets, according to the generic CSS grammar and the individual grammars of each property in this module.

CR exit criteria

As described in the W3C process document, a Candidate Recommendation (CR) is a specification that W3C recommends for use on the Web. The next stage is "Recommendation" when the specification is sufficiently implemented.

For this specification to be proposed as a W3C Recommendation, the following conditions shall be met. There must be at least two independent, interoperable implementations of each feature. Each feature may be implemented by a different set of products, there is no requirement that all features be implemented by a single product. For the purposes of this criterion, we define the following terms:

independent
each implementation must be developed by a different party and cannot share, reuse, or derive from code used by another qualifying implementation. Sections of code that have no bearing on the implementation of this specification are exempt from this requirement.
interoperable
passing the respective test case(s) in the official CSS test suite, or, if the implementation is not a Web browser, an equivalent test. Every relevant test in the test suite should have an equivalent test created if such a user agent (UA) is to be used to claim interoperability. In addition if such a UA is to be used to claim interoperability, then there must one or more additional UAs which can also pass those equivalent tests in the same way for the purpose of interoperability. The equivalent tests must be made publicly available for the purposes of peer review.
implementation
a user agent which:
  1. implements the specification.
  2. is available to the general public. The implementation may be a shipping product or other publicly available version (i.e., beta version, preview release, or "nightly build"). Non-shipping product releases must have implemented the feature(s) for a period of at least one month in order to demonstrate stability.
  3. is not experimental (i.e., a version specifically designed to pass the test suite and is not intended for normal usage going forward).

A minimum of sixth months of the CR period must have elapsed. This is to ensure that enough time is given for any remaining major errors to be caught.

Features will be dropped if two or more interoperable implementations are not found by the end of the CR period.

Features may/will also be dropped if adequate/sufficient (by judgment of CSS WG) tests have not been produced for those feature(s) by the end of the CR period.

Appendix E — Acknowledgements

The editors would like to thank the members of the W3C Voice Browser and Cascading Style Sheets working groups for their assistance in preparing this new draft. Special thanks to Ellen Eide (IBM) for her detailed comments.

Appendix F — Changes from previous draft

Appendix G — References

Normative references

[CSS3BOX]
Bert Bos. CSS basic box model. 9 August 2007. W3C Working Draft. (Work in progress.) URL: http://www.w3.org/TR/2007/WD-css3-box-20070809
[CSS3VAL]
Håkon Wium Lie; Chris Lilley. CSS3 Values and Units. 19 September 2006. W3C Working Draft. (Work in progress.) URL: http://www.w3.org/TR/2006/WD-css3-values-20060919
[RFC2119]
S. Bradner. Key words for use in RFCs to Indicate Requirement Levels. Internet RFC 2119. URL: http://www.ietf.org/rfc/rfc2119.txt
[SPEECH-SYNTHESIS]
Daniel C. Burnett; Mark R. Walker; Andrew Hunt. Speech Synthesis Markup Language (SSML) Version 1.0. 7 September 2004. W3C Recommendation. URL: http://www.w3.org/TR/2004/REC-speech-synthesis-20040907/
[XML10]
C. M. Sperberg-McQueen; et al. Extensible Markup Language (XML) 1.0 (Fifth Edition). 26 November 2008. W3C Recommendation. URL: http://www.w3.org/TR/2008/REC-xml-20081126/
[XML11]
Eve Maler; et al. Extensible Markup Language (XML) 1.1 (Second Edition). 16 August 2006. W3C Recommendation. URL: http://www.w3.org/TR/2006/REC-xml11-20060816
[IPA]
International Phonetic Association

Other references

[CSS21]
Bert Bos; et al. Cascading Style Sheets Level 2 Revision 1 (CSS 2.1) Specification. 23 April 2009. W3C Candidate Recommendation. (Work in progress.) URL: http://www.w3.org/TR/2009/CR-CSS2-20090423
[CSS3CASCADE]
Håkon Wium Lie. CSS3 module: Cascading and inheritance. 15 December 2005. W3C Working Draft. (Work in progress.) URL: http://www.w3.org/TR/2005/WD-css3-cascade-20051215
[CSS3GENCON]
Ian Hickson. CSS3 Generated and Replaced Content Module. 14 May 2003. W3C Working Draft. (Work in progress.) URL: http://www.w3.org/TR/2003/WD-css3-content-20030514
[CSS3SYN]
L. David Baron. CSS3 module: Syntax. 13 August 2003. W3C Working Draft. (Work in progress.) URL: http://www.w3.org/TR/2003/WD-css3-syntax-20030813