[opus] Opus voice activity detector

daniel.brox at picotera.co daniel.brox at picotera.co
Thu Nov 5 23:34:48 UTC 2020


Trying to implement a classic spectral subtraction algorithm using Opus
voice activity detector, and have a couple questions: 

1.Should there be any overlap between PCM frames input to the voice
activity detector, and if so, what windowing (e.g. Hamming) should be
applied to the frames before being input? 

2.Is there a threshold value for either the speech activity level of a
frame or the probability of speech occurrence that is typically used to
assert that speech did or did not occur in the frame? 

Thanks
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.xiph.org/pipermail/opus/attachments/20201105/3e4a0f38/attachment.html>


More information about the opus mailing list