Re: GNOME 2 Sound Architecture and APIs?

[Bill Haneman <bill haneman ireland sun com>]

Elliot Lee wrote:
<snip> 
> > text-to-speech will require of the sound subsystem:
> >
> > latency <= 100 ms
> > ability to stop sound emission
> > ability to readily mix sound requests to play concurrently
> 
> I think the existing esound solution can do #3 fine, and #1 not horribly
> far off. AFAICS #2 is mostly a problem for the application - if you don't
> want sound to be output, don't output it!

Well, I was referring to the case of "here's a sound to play"..."no,
wait, stop playing that".  

Screen readers do this all the time with text-to-speech; as a user
tabs down a menu there are repeated requests to speak the text,
followed immediately by requests to stop speaking the previous and
start the new text...

Since the time it takes to utter a label is much longer than the time
it takes to tab from widget to widget, if you don't interrupt the
previous utterance you will end up with way too many overlapping
utterances.

-Bill


-- 
--------------
Bill Haneman
Gnome Accessibility / Batik SVG Toolkit
Sun Microsystems Ireland