W3C home > Mailing lists > Public > www-style@w3.org > September 2011

Re: [css3-speech] Volume properties

From: Daniel Weck <daniel.weck@gmail.com>
Date: Thu, 8 Sep 2011 17:45:21 +0100
Message-Id: <026540DD-94C7-4850-9A0B-40A57CB76FCD@gmail.com>
To: Christoph Päper <christoph.paeper@crissov.de>, W3C style mailing list <www-style@w3.org>
Hello Christoph,
thank you for your ideas about the next generation of CSS Speech (Level 4+). That's certainly good food for thoughts, useful for when 3D spatial audio will be discussed in Level 4.

With regards to your suggestion to rename 'voice-volume' to 'volume-voice': I think you are making the premature assumption that a shorthand value will naturally cover the needs of both Text-To-Speech (which is the scope of the current CSS Speech specification) and more generic multimedia control (e.g. CSS properties to control embedded document audio/video, either authored in markup or injected via the "content:" predicate). I think that the dichotomy between these two domains will remain. The 'cue' feature (pre-recorded audio clips) in CSS Speech is very specific to how a screen-reader-like aural experience is structured (not limited to accessibility needs, as you know), thus why this is encapsulated within the "aural box model" right now. You mention future consistency as an argument, but I actually think that consistency in the current Level 3 context is important, so I am not in favor of renaming 'voice-volume'.

Let us know if this is a satisfactory answer.
Regards, Daniel

On 19 Aug 2011, at 14:07, Christoph Päper wrote:

> Bert Bos:
> 
>>   http://www.w3.org/TR/2011/WD-css3-speech-20110818/
> 
> The introductionary section 4 really helps CSS-savvy people to quickly grasp essential concepts, thanks.
> 
> Of course, only one dimension of the 2D visual box model – it’s 2½D if you count in ‘z-index’ and table layout – can map to the 1D aural box model. The editors chose the vertical domain (‘top’/ [‘middle’] / ‘bottom’) for the diagram over the horizontal (‘right’ / [‘center’] / ‘left’) and logical one (‘start’ / ‘end’), although they use the second set of logical terms (‘before’ / ‘after’), which makes sense, because we do only have ‘::before’ and ‘::after’ for runs of text, which are also 1D (although their consituents are 2D in visual media).
> 
> Hm, that became more a note to myself. Originally, I wanted to suggest to use ‘margin-before’ etc. instead, if that was agreed upon. A future Box module should cover all models and dimensions:
> * 1D temporal                – t
> * 1D linear                  – x
> * 1½D temporal layers        – tv
> * 1½D linear layers          – xv
> * 2D planar                  – xy
> * 2D marquee                 – tx
> * 2½D planar layers          – xyv
> * 2½D marquee layers         – txv
> * 3D spatial                 – xyz
> * 3½D temporal-planar layers – txyv
> * 4D temporal-spatial        – txyz
> 
> The Speech module currently is “t”, but should probably become “tv” in the future, see below.
> 
>> The module contains the properties to style how a document is spoken by 
>> a speech synthesizer: voice, volume, speed, pauses, cue sounds, etc.
> 
> I believe in level 4 this module should be split into two, separating aural and speech (synthesis) properties, because one might want to do, following Example XIII,
> 
>  @media aural { article#poem {
>    content: url(Poem_performance.audio);
>    audio-volume: soft;
>  }}
>  @media speech { article#poem {
>    background: url(ambience.audio);
>    volume: loud /* speech */ x-soft /* background */;
>  }}
> 
> or
> 
>  …
>    voice-volume: loud;
>    voice-opacity: 80% /* or 0.8 */;
>  }}
> 
> Therefore I think ‘volume-voice’ makes more sense than ‘voice-volume’, because then we can more consistently introduce a shorthand property ‘volume’ later.
> 
> PS: You could extend that example for other media
> 
>  @media print {article#poem {
>    content: url(Poem_calligraphy.image);
>  }}
>  @media projection {article#poem {
>    content: url(Poem_illustrated.animation);
>  }}
>  @media tv {article#poem {
>    content: url(Poem_illustrated_performance.video);
>  }}
>  @media screen {article#poem {
>    content: url(Poem_experience.app);
>  }}
> 
> 
Received on Thursday, 8 September 2011 16:45:54 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:44 GMT