[Speex-dev] Behavior of Speex VAD algorithm

[Meir Yeshurun]

Hi everyone!

I have two questions regarding the Speex VAD mechanism:

1) I noticed a delay of about 0.4 seconds in recognizing the end of speech 
(the point where there is a transition from speech to silence). Is there 
some way to prevent this delay?

2) When the recording immediately starts with voice rather than silence, 
Speex is confused for a few seconds, and usually only recovers after the 
first interval of speech ends. Is there someway to prevent this problem? I 
thought of prefixing each stream with a short recording of silence. Is there 
a better way?

Thanks for your help!

_________________________________________________________________
Don't just search. Find. Check out the new MSN Search! 
http://search.msn.com/