Skip to content

Commit 95e53fc

Browse files
danielweckdanielweck
committed
updated draft with list-issue, fixed audio-cues volume level
1 parent 1fa191e commit 95e53fc

File tree

2 files changed

+169
-115
lines changed

2 files changed

+169
-115
lines changed

‎css3-speech/Overview.html‎

Lines changed: 116 additions & 86 deletions
Original file line numberDiff line numberDiff line change
@@ -5,7 +5,8 @@
55
<head>
66
<title>CSS3 Speech Module</title>
77
<meta content="text/html; charset=us-ascii" http-equiv=Content-Type>
8-
<link href=default.css rel=stylesheet type="text/css">
8+
<link href="http://dev.w3.org/csswg/default.css" rel=stylesheet
9+
type="text/css">
910

1011
<style type="text/css">
1112
.prod { font-family: inherit; font-size: inherit }
@@ -28,7 +29,8 @@
2829
}
2930
/*]]>*/
3031
</style>
31-
<link href="http://www.w3.org/StyleSheets/TR/W3C-ED.css" rel=stylesheet
32+
<!-- link rel="stylesheet" type="text/css" href="http://www.w3.org/StyleSheets/TR/W3C-ED.css" -->
33+
<link href="http://www.w3.org/StyleSheets/TR/W3C-ED" rel=stylesheet
3234
type="text/css">
3335

3436
<body>
@@ -38,13 +40,12 @@
3840

3941
<h1 id=top>CSS3 Speech Module</h1>
4042

41-
<h2 class="no-num no-toc" id=longstatus-date>Editor's Draft 07 February
42-
2011</h2>
43+
<h2 class="no-num no-toc" id=longstatus-date>Editor's Draft 05 April 2011</h2>
4344

4445
<dl>
4546
<dt>This version:
4647

47-
<dd><!--<a href="http://www.w3.org/TR/2011/WD-css3-speech-20110207">http://www.w3.org/TR/2011/ED-css3-speech-20110207</a>-->
48+
<dd><!--<a href="http://www.w3.org/TR/2011/WD-css3-speech-20110405">http://www.w3.org/TR/2011/ED-css3-speech-20110405</a>-->
4849
<a
4950
href="http://dev.w3.org/csswg/css3-speech">http://dev.w3.org/csswg/css3-speech</a>
5051

@@ -158,20 +159,32 @@ <h2 class="no-num no-toc" id=status>Status of this document</h2>
158159
href="/Voice/Group/">Voice Browser working group</a> (<a
159160
href="http://cgi.w3.org/MemberAccess/AccessRequest">W3C Members only</a>).
160161

161-
<p>The following features are at-risk and may be dropped at the end of the
162-
CR period if there has not been enough interest from implementers:
162+
<p>&nbsp;</p>
163+
<!-- p>The following features are judged "at risk" and may be dropped at the
164+
end of the CR period,
165+
especially if there has not been enough interest from implementers:
166+
<ul>
167+
168+
</ul>
169+
</p -->
170+
<!-- li><a href="#issue-mark">mark</a></li -->
171+
172+
<p>The following issues need to be discussed and require working group
173+
resolutions:
163174

164175
<ul>
165-
<li><a href="#issue-phonemes">phonemes</a>
176+
<li><a href="#issue-phonemes">&lsquo;<code
177+
class=property>phonemes</code>&rsquo; property breaks principle of
178+
content/presentation separation</a>
166179

167-
<li><a href="#issue-voice-cues-volume">audio cues volume level relative to
168-
voice-volume</a>
180+
<li><a href="#issue-speakability-name">&lsquo;<code
181+
class=property>speakability</code>&rsquo; not an ideal property name
182+
?</a>
169183

170-
<li><a href="#issue-speakability-name">"speakability" not an ideal name
171-
?</a></li>
172-
<!-- li><a href="#issue-mark">mark</a></li -->
173-
<!-- li><a href="#issue-speak_none">speak:none</a></li -->
184+
<li><a href="#issue-lists">handling of list styles, numbered items</a>
174185
</ul>
186+
<!-- li><a href="#issue-voice-cues-volume">audio cues volume level relative to voice-volume</a></li -->
187+
<!-- li><a href="#issue-speak_none">speak:none</a></li -->
175188

176189
<p>The CSS WG maintains an <a
177190
href="http://www.w3.org/Style/CSS/Tracker/products/29">issues list</a> for
@@ -462,23 +475,19 @@ <h2 id=mixing-props><span class=secno>4. </span>Mixing properties:
462475
class=property>voice-volume</code></a>&rsquo; refers to the amplitude of
463476
the waveform output by the speech synthesiser. This may be mixed with
464477
other audio sources, influencing the perceived loudness of synthetic
465-
speech relative to these sources.
466-
467-
<p> &lsquo;<a href="#voice-volume"><code
468-
class=property>voice-volume</code></a>&rsquo; does not apply to <a
469-
href="#cue-props">audio cues</a> for which there is a separate means to
470-
set loudness (absolute, or relative to &lsquo;<a
471-
href="#voice-volume"><code class=property>voice-volume</code></a>&rsquo;).
472-
<span class=issue id=issue-voice-cues-volume> ISSUE: shouldn't audio cues
473-
be subjected to the voice-volume, so that an inaudible element gets its
474-
entire content (within the aural "box" model) muted as well ? The
475-
rationale is that voice-volume:silent is largely equivalent to
476-
visibility:hidden, which affects border, padding and spacing (i.e. the
477-
entire content with regards to the CSS "visual" box model). The audio cues
478-
should therefore only be allowed to set their volume levels
479-
<em>relatively</em> to the inherited voice-volume, not with absolute
480-
values as these may incorrectly enable audio output when the voice-volume
481-
itself is muted. </span>
478+
speech relative to these sources.</p>
479+
<!-- p>
480+
&lsquo;<code class=property>voice-volume</code>&rsquo; does not apply to <a href
481+
="#cue-props">audio cues</a> for which there is a separate means
482+
to set loudness (absolute, or relative to &lsquo;<code class=property>voice-volume</code>&rsquo;).
483+
<span class="issue" id="issue-voice-cues-volume">
484+
ISSUE: shouldn't audio cues be subjected to the voice-volume,
485+
so that an inaudible element gets its entire content (within the aural "box" model) muted as well ?
486+
The rationale is that voice-volume:silent is largely equivalent to visibility:hidden,
487+
which affects border, padding and spacing (i.e. the entire content with regards to the CSS "visual" box model).
488+
The audio cues should therefore only be allowed to set their volume levels <em>relatively</em> to the inherited voice-volume, not with absolute values as these may incorrectly enable audio output when the voice-volume itself is muted.
489+
</span>
490+
</p -->
482491

483492
<p>Values have the following meanings:</p>
484493
<!-- dl>
@@ -637,11 +646,9 @@ <h2 id=mixing-props><span class=secno>4. </span>Mixing properties:
637646
<p>Many speech synthesizers only support a single channel. The &lsquo;<a
638647
href="#voice-balance"><code class=property>voice-balance</code></a>&rsquo;
639648
property can then be treated as part of a post synthesis mixing step. This
640-
is where speech is mixed with other audio sources. Unlike &lsquo;<a
641-
href="#voice-volume"><code class=property>voice-volume</code></a>&rsquo;,
642-
&lsquo;<a href="#voice-balance"><code
643-
class=property>voice-balance</code></a>&rsquo; <em>does</em> apply to <a
644-
href="#cue-props">audio cues</a>.
649+
is where speech is mixed with other audio sources.</p>
650+
<!-- p>Unlike &lsquo;<code class=property>voice-volume</code>&rsquo;, &lsquo;<code class=property>voice-balance</code>&rsquo;
651+
<em>does</em> apply to <a href="#cue-props">audio cues</a>.</p -->
645652

646653
<h2 id=speaking-props><span class=secno>5. </span>Speaking properties:
647654
&lsquo;<a href="#speakability"><code
@@ -1282,10 +1289,11 @@ <h2 id=cue-props><span class=secno>8. </span>Cue properties: &lsquo;<a
12821289
<td><dfn id=cue-before>cue-before</dfn>
12831290

12841291
<tr>
1285-
<td width="20%"><em>Value:</em>
1292+
<td width="20%"><em>Value:</em></td>
1293+
<!-- td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; | silent
1294+
| x-soft | soft | medium | loud | x-loud] | none | inherit</td -->
12861295

1287-
<td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; |
1288-
silent | x-soft | soft | medium | loud | x-loud] | none | inherit
1296+
<td>&lt;uri&gt; [&lt;percentage&gt;] | none | inherit
12891297

12901298
<tr>
12911299
<td width="20%"><em>Initial:</em>
@@ -1324,10 +1332,11 @@ <h2 id=cue-props><span class=secno>8. </span>Cue properties: &lsquo;<a
13241332
<td><dfn id=cue-after>cue-after</dfn>
13251333

13261334
<tr>
1327-
<td width="20%"><em>Value:</em>
1335+
<td width="20%"><em>Value:</em></td>
1336+
<!-- td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; | silent
1337+
| x-soft | soft | medium | loud | x-loud] | none | inherit</td -->
13281338

1329-
<td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; |
1330-
silent | x-soft | soft | medium | loud | x-loud] | none | inherit
1339+
<td>&lt;uri&gt; [&lt;percentage&gt;] | none | inherit
13311340

13321341
<tr>
13331342
<td width="20%"><em>Initial:</em>
@@ -1371,43 +1380,44 @@ <h2 id=cue-props><span class=secno>8. </span>Cue properties: &lsquo;<a
13711380

13721381
<dt><strong>none</strong>
13731382

1374-
<dd>No auditory icon is specified.
1375-
1376-
<dt><strong>&lt;non-negative number&gt;</strong>
1383+
<dd>No auditory icon is specified.</dd>
1384+
<!-- dt><strong>&lt;non-negative number&gt;</strong></dt>
13771385
1378-
<dd>An integer or floating point number in the range &lsquo;<code
1379-
class=css>0</code>&rsquo; to &lsquo;<code class=css>100</code>&rsquo;.
1380-
&lsquo;<code class=css>0</code>&rsquo; represents silence (the
1381-
<em>minimum</em> level), and 100 corresponds to the <em>maximum</em>
1382-
level. The volume scale is linear amplitude.
1386+
<dd>An integer or floating point number in the range &lsquo;<code class=css>0</code>&rsquo; to &lsquo;<code class=css>100</code>&rsquo;.
1387+
&lsquo;<code class=css>0</code>&rsquo; represents silence (the <em>minimum</em> level), and 100
1388+
corresponds to the <em>maximum</em> level. The volume scale is
1389+
linear amplitude.</dd -->
13831390

13841391
<dt><strong>&lt;percentage&gt;</strong>
13851392

13861393
<dd> Only positive values are allowed. Computed values are calculated
13871394
relative to the inherited value of the &lsquo;<a
13881395
href="#voice-volume"><code class=property>voice-volume</code></a>&rsquo;
13891396
property, and are then clipped to the range &lsquo;<code
1390-
class=css>0</code>&rsquo; to &lsquo;<code class=css>100</code>&rsquo;. By
1391-
basing the percentage upon the inherited value for &lsquo;<a
1392-
href="#voice-volume"><code class=property>voice-volume</code></a>&rsquo;,
1393-
it is easy to adjust the relative loudness of cues compared to synthetic
1394-
speech for whatever volume setting has been provided for that speech.
1397+
class=css>0</code>&rsquo; to &lsquo;<code class=css>100</code>&rsquo;.
1398+
&lsquo;<code class=css>0</code>&rsquo; represents silence (the
1399+
<em>minimum</em> level), and 100 corresponds to the <em>maximum</em>
1400+
level. The volume scale is linear amplitude. By basing the percentage
1401+
upon the inherited value for &lsquo;<a href="#voice-volume"><code
1402+
class=property>voice-volume</code></a>&rsquo;, it is easy to adjust the
1403+
relative loudness of cues compared to synthetic speech for whatever
1404+
volume setting has been provided for that speech.
13951405
<p class=note> Note that a leading "+" sign does not denote an increment.
13961406
For example, +50% is equivalent to 50%, so the computed value equals the
13971407
inherited value times 0.5 (divided by 2), then clipped to [0,100].</p>
1398-
1399-
<dt><strong>silent</strong>, <strong>x-soft</strong>,
1400-
<strong>soft</strong>, <strong>medium</strong>, <strong>loud</strong>,
1401-
and <strong>x-loud</strong>
1402-
1403-
<dd>A sequence of monotonically non-decreasing volume levels. The value of
1404-
&lsquo;<code class=property>silent</code>&rsquo; is mapped to
1405-
&lsquo;<code class=css>0</code>&rsquo; and &lsquo;<code
1406-
class=property>x-loud</code>&rsquo; is mapped to &lsquo;<code
1407-
class=css>100</code>&rsquo;. The mapping of other values to numerical
1408-
volume levels is implementation dependent, but the intention is to match
1409-
the corresponding levels for &lsquo;<a href="#voice-volume"><code
1410-
class=property>voice-volume</code></a>&rsquo;.
1408+
</dd>
1409+
<!-- dt><strong>silent</strong>,
1410+
<strong>x-soft</strong>,
1411+
<strong>soft</strong>,
1412+
<strong>medium</strong>,
1413+
<strong>loud</strong>, and
1414+
<strong>x-loud</strong></dt>
1415+
1416+
<dd>A sequence of monotonically non-decreasing volume levels.
1417+
The value of &lsquo;<code class=property>silent</code>&rsquo; is mapped to &lsquo;<code class=css>0</code>&rsquo; and &lsquo;<code class=property>x-loud</code>&rsquo; is mapped
1418+
to &lsquo;<code class=css>100</code>&rsquo;. The mapping of other values to numerical volume levels
1419+
is implementation dependent, but the intention is to match the
1420+
corresponding levels for &lsquo;<code class=property>voice-volume</code>&rsquo;.</dd-->
14111421
</dl>
14121422

14131423
<div class=example>
@@ -1416,9 +1426,9 @@ <h2 id=cue-props><span class=secno>8. </span>Cue properties: &lsquo;<a
14161426
<pre>
14171427
a { cue-before: url(bell.aiff); cue-after: url(dong.wav) }
14181428

1419-
h1 { cue-before: url(pop.au) 80; cue-after: url(pop.au) 50% }
1429+
h1 { cue-before: url(pop.au) 80%; cue-after: url(pop.au) 50% }
14201430

1421-
div.caution { cue-before: url(caution.wav) loud }
1431+
div.caution { cue-before: url(caution.wav) 130% }
14221432
</pre>
14231433
</div>
14241434

@@ -2206,8 +2216,15 @@ <h2 id=phonetic-props><span class=secno>11. </span>Phonetics: &lsquo;<a
22062216

22072217
<p class=issue id=issue-phonemes>ISSUE: the &lsquo;<a
22082218
href="#phonemes"><code class=property>phonemes</code></a>&rsquo; property
2209-
covers functionality that is arguably not strictly related to styling (it
2210-
relates to the broader aspect of pronunciation lexicons). There is a <a
2219+
covers functionality that arguably doesn't belong to CSS, as it breaks the
2220+
principle of separation between content and presentation (i.e. any change
2221+
in the source text must be echoed by an corresponding change in the
2222+
content of the aural &lsquo;<a href="#phonemes"><code
2223+
class=property>phonemes</code></a>&rsquo; property). The broader aspect of
2224+
pronunciation lexicons would be better addressed in the content markup.
2225+
For example, <a href="http://wiki.whatwg.org/wiki/RelExtensions">it is
2226+
proposed</a> that formats like the W3C's own PLS standard should be
2227+
supported by the link/rel mechanism. There is a <a
22112228
href="http://lists.w3.org/Archives/Public/www-style/2011Jan/0248.html">related
22122229
discussion</a> on the public mailing-list.
22132230

@@ -2320,9 +2337,9 @@ <h2 id=content><span class=secno>12. </span>Inserted and replaced content
23202337
the text-to-speech with its own synthesizer.</p>
23212338
</div>
23222339

2323-
<p>Furthermore authors, or users in a user stylesheet, may want to add some
2324-
information to ease understanding the structure for non-visual interaction
2325-
with the document. They can do so by using the &lsquo;<code
2340+
<p>Furthermore, authors (or users in a user stylesheet) may want to add
2341+
some information to ease understanding the structure for non-visual
2342+
interaction with the document. They can do so by using the &lsquo;<code
23262343
class=css>::before</code>&rsquo; and &lsquo;<code
23272344
class=css>::after</code>&rsquo; pseudo-elements that will be inserted
23282345
between the element's contents and the &lsquo;<a href="#rest"><code
@@ -2347,7 +2364,17 @@ <h2 id=content><span class=secno>12. </span>Inserted and replaced content
23472364

23482365
<p class=note>Note that detailed information can be found in the CSS3
23492366
Generated and Replaced Content Module <a href="#CSS3GENCON"
2350-
rel=biblioentry>[CSS3GENCON]<!--{{CSS3GENCON}}--></a>.</p>
2367+
rel=biblioentry>[CSS3GENCON]<!--{{CSS3GENCON}}--></a>.
2368+
2369+
<p class=issue id=issue-lists> ISSUE: the speech handling of list items is
2370+
under-specified. One suggestion is to ignore list-style-type (which can
2371+
represent visual glyphs not suitable for aural processing anyway) and to
2372+
define a standard way to read <a
2373+
href="http://dev.w3.org/csswg/css3-lists/#marker-pseudoelement">list
2374+
"markers"</a>. Unfortunately this introduces a dependency on a feature
2375+
specific to CSS3-Lists. There is a <a
2376+
href="http://lists.w3.org/Archives/Public/www-style/2011Feb/0195.html">related
2377+
discussion</a> on the public mailing-list.</p>
23512378
<!--
23522379
<h2 id="interpret-as-props">Interpretation property:
23532380
<a href="#interpret-as">&lsquo;<code class=property>interpret-as</code>&rsquo;</a></h2>
@@ -2486,8 +2513,7 @@ <h2 class=no-num id=property-index>Appendix B &mdash; Property index</h2>
24862513
<tr valign=baseline>
24872514
<td><a class=property href="#cue-after">cue-after</a>
24882515

2489-
<td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; |
2490-
silent | x-soft | soft | medium | loud | x-loud] | none | inherit
2516+
<td>&lt;uri&gt; [&lt;percentage&gt;] | none | inherit
24912517

24922518
<td>none
24932519

@@ -2502,8 +2528,7 @@ <h2 class=no-num id=property-index>Appendix B &mdash; Property index</h2>
25022528
<tr valign=baseline>
25032529
<td><a class=property href="#cue-before">cue-before</a>
25042530

2505-
<td>&lt;uri&gt; [&lt;non-negative number&gt; | &lt;percentage&gt; |
2506-
silent | x-soft | soft | medium | loud | x-loud] | none | inherit
2531+
<td>&lt;uri&gt; [&lt;percentage&gt;] | none | inherit
25072532

25082533
<td>none
25092534

@@ -3116,8 +3141,13 @@ <h2 class=no-num id=changes>Appendix F &mdash; Changes from previous draft</h2>
31163141
<li>Fixed &lsquo;<a href="#voice-family"><code
31173142
class=property>voice-family</code></a>&rsquo; grammar as per <a
31183143
href="http://lists.w3.org/Archives/Public/www-style/2010Dec/0231.html">this
3119-
discussion</a></li>
3120-
<!-- li>Added the default &lsquo;<code class=property>auto</code>&rsquo; value for the &lsquo;<code class=property>voice-volume</code>&rsquo; property, in order to inherit aural rendering behavior from the &lsquo;<code class=property>visibility</code>&rsquo; property.</li -->
3144+
discussion</a>
3145+
3146+
<li>The volume level of audio cues can only be set relatively to the
3147+
inherited &lsquo;<a href="#voice-volume"><code
3148+
class=property>voice-volume</code></a>&rsquo; property (to avoid cues
3149+
being spoken when the main element is silent, which contradicts the
3150+
"aural box model").
31213151

31223152
<li>Added "HTML" to "CSS defines aural properties that give control over
31233153
rendering XML to speech" in the abstract.
@@ -3252,10 +3282,10 @@ <h3 class=no-num id=other-references>Other references</h3>
32523282
<dt id=CSS21>[CSS21]
32533283

32543284
<dd>Bert Bos; et al. <a
3255-
href="http://www.w3.org/TR/2009/CR-CSS2-20090908"><cite>Cascading Style
3256-
Sheets Level 2 Revision 1 (CSS 2.1) Specification.</cite></a> 8 September
3257-
2009. W3C Candidate Recommendation. (Work in progress.) URL: <a
3258-
href="http://www.w3.org/TR/2009/CR-CSS2-20090908">http://www.w3.org/TR/2009/CR-CSS2-20090908</a>
3285+
href="http://www.w3.org/TR/2010/WD-CSS2-20101207"><cite>Cascading Style
3286+
Sheets Level 2 Revision 1 (CSS 2.1) Specification.</cite></a> 7 December
3287+
2010. W3C Working Draft. (Work in progress.) URL: <a
3288+
href="http://www.w3.org/TR/2010/WD-CSS2-20101207">http://www.w3.org/TR/2010/WD-CSS2-20101207</a>
32593289
</dd>
32603290
<!---->
32613291

0 commit comments

Comments
 (0)