Comments on min/max for speech synthesizer voice characteristics

Hello,

This morning, Jon and I discussed the current efforts of the
Working Group [1] to identify min/max values for various speech
synthesizer parameters: speech rate, pitch, gender, pitch range,
stress, and richness. Jon made the point (as he has in email [2])
that values for various parameters are usually attained through
communication between the user agent and the speech engine
(which I take to be software, hardware, or both).

Based on Jon's comments, I propose the following:

 1) We do not put minimal requirements for speech parameters
    in the Guidelines, but instead state how the range of values for
    the parameters in question is determined. In Jon's email [2],
    for example, he lists the parameters available through the SAPI
    interface [3]. (Is there HTML documentation for SAPI on the Web,
    or do I have to download software or a big document to read
    about the interface?) 

    I do not have a proposal for how to express this communication,
    and welcome input from those versed in speech synthesizer
    technology.ww

 2) We should put some example ranges in the Techniques document.
    Thus, the min/max range for speech rate of 100 to 700 
    words/minute would be listed in Techniques, as well as any
    other data people submit.

 3) We refer developers also to CSS aural style sheets.

 - Ian


[1] http://lists.w3.org/Archives/Public/w3c-wai-ua/2000JulSep/0000.html
[2] http://lists.w3.org/Archives/Public/w3c-wai-ua/2000JulSep/0002.html
[3] http://www.microsoft.com/IIT/documentation/online.htm
-- 
Ian Jacobs (jacobs@w3.org)   http://www.w3.org/People/Jacobs
Tel:                         +1 831 457-2842
Cell:                        +1 917 450-8783

Received on Monday, 3 July 2000 17:54:52 UTC