Chunking speech streams by words into sounds that pygame can play on Vista
fails miserably because of some pygame bug. It also results in very choppy
speech on XP and Linux. Since we know the sample rate, we can use a timing
based approach to word callbacks that should be accurate enough for doing
any kind of tracking in the browser. Start and end will still be fired as
they occur. On end, all unfired word callbacks will be fired ahead of the
end event.
Original issue reported on code.google.com by c...@unc.edu on 5 Sep 2008 at 10:53
Original issue reported on code.google.com by
c...@unc.edu
on 5 Sep 2008 at 10:53