Re: [css3-speech] LC comment: voice-rate percentages

Hi again Greg, here is the proposed prose improvement:

-----
<percentage>
Only non-negative percentage values are allowed. This represents a change relative to the given keyword value (see enumeration above), or to the default value for the root element, or otherwise to the inherited speaking rate (which may itself be a combination of a keyword value and of a percentage, in which case percentages are combined multiplicatively).

For example: 50% means that the speech rate gets multiplied by 0.5 (the speed is divided by two). 200% means that the rate gets multiplied by 2 (the speech is twice as fast). 150% corresponds to a 50% increase (the base rate is multiplied by 1.5).
To summarize, the speech rate remains unchanged when the given percentage is 100%, the rate is decreased with percentages between 0 and 100, and the rate is increased with percentages above 100.
-----

Cheers, Daniel

On 20 Oct 2011, at 00:23, Daniel Weck wrote:

> Hello Greg
> just a heads-up to let you know that the CSS Working Group has reviewed this issue, and the consensus is to make a minor editorial improvement in order to clarify the effect of percentages, using more common terms such as "slower" and "faster". Regarding the authoring guidelines and good practices, the CSS Working Group agrees that this is out-of-scope as far as the CSS Speech Module is concerned. However, note that I started to track-down a number of cross-cutting concerns that will need to be discussed in collaboration with other groups (including WAI). To be continued, watch this space.
> 
> As per the W3C process, you may choose to accept this resolution, or you may decide to raise an objection. Please let us know, and thanks again for your valuable input regarding screen-reader usage.
> 
> Kind regards, Dan
> 
> On 12 Oct 2011, at 14:55, Daniel Weck wrote:
> 
>> Hi Greg,
>> 
>> Regarding "COMMENT 1": I re-read the current specification prose, I re-read your suggested editorial changes, and I came to the conclusion that the current prose already addresses your concerns. We explicitly give an example for 50% (half the inherited speaking rate), and further below we even give an example for 120%. We also express the fact that the baseline rate to which the percentage applies is speech processor-dependent. However, there is a mistake/typo in the example: "multiplied by 1.2" is not "one and a half times the speaking rate". This needs to be fixed.
>> 
>> Regarding "COMMENT 2": I don't think it is the role of a CSS specification to provide authoring guidelines (even informatively). I will double-check with the more seasoned Working Group members, but I believe it is uncommon practice to advise authors in the way you suggest.
>> 
>> Please let us know whether this response is satisfactory.
>> Kind regards, Dan
>> 
>> On 1 Oct 2011, at 04:13, Gregory Rosmaita wrote:
>> 
>>> aloha!
>>> 
>>> in regards voice-rate percentage values, the LC draft of css3-speech
>>> states:
>>> 
>>> QUOTE
>>> <percentage>
>>>        Only non-negative percentage values are allowed. This
>>>        represents a change relative to the given keyword value (see
>>>        enumeration above), or to the default value for the root
>>>        element, or otherwise to the inherited speaking rate (which may
>>>        itself be a combination of a keyword value and of a percentage,
>>>        in which case percentages are combined multiplicatively). For
>>>        example, 50% means that the speaking rate gets multiplied by
>>>        0.5 (half the value).
>>> UNQUOTE
>>> 
>>> COMMENT 1: if the "currently active" voice rate is represented by 100%,
>>> and from that "baseline" scaled up or down (depending upon whether one
>>> wants to increase or deecrease the rate) using positive percentage
>>> values ONLY, then that needs to be explicitly stated in the
>>> css3-speech recommendation.  while the 100% "baseline" concept is far
>>> from novel to those proficient with CSS, it is not an obvious or "self
>>> evident" convention, and therefore should be explicitly stated so as
>>> to eliminate confusion as to how positive percentages greater than
>>> 100% can be used to increase voice-rate and how positive percentages
>>> less than 100%, but greater than 0%, are used to decrease voice-rate.
>>> 
>>> SUGGESTED TEXT:
>>> 
>>> A value less than 100% slows down the voice-rate. Values of greater
>>> than 100% indicate an increase in voice-rate.  The actual rate of
>>> speech relative to the "currently active rate" of 100% is determined
>>> by the capacities of the speech engine being used.
>>> 
>>> 
>>> COMMENT 2. as an author and end user, i would be most comfortable using
>>> "voice-rate" to effect small-scale changes in voice rate that are
>>> neither disorienting nor painful for the end user to aurally process:
>>> 
>>> body { voice-rate: inherit;
>>>    code { voice-rate: 90%; voice-stress: none;
>>> 	     speak-as: literal-punctuation; }
>>>    em { voice-rate: 110%; voice-stress: moderate; }
>>>    strong { voice-rate: 115%; voice-stress: strong; }
>>> 
>>> authors should also be cautioned against using rate and/or volume alone
>>> or in tandem to indicate a specific type of markup, as control over the
>>> rate and volume of speech and its relative values are EXTREMELY important
>>> user-defined settings, the bounds of which which are based upon the user's
>>> needs, experience, other abilities/disabilities, and the type of content
>>> being converted into speech.
>>> 
>>> thanks, gregory.
>>> 
>>> ----------------------------------------------------------
>>> ACCOUNTABILITY, n. The mother of caution.
>>>               -- Ambrose Bierce, The Devil's Dictionary
>>> ----------------------------------------------------------
>>>   Gregory J. Rosmaita, gregory.rosmaita@gmail.com
>>>    Camera Obscura: http://www.hicom.net/~oedipus/
>>> Oedipus' Online Complex: http://my.opera.com/oedipus/
>>> ----------------------------------------------------------
>>> 
>> 
>

Received on Thursday, 20 October 2011 01:25:10 UTC