I 'd like understand how VAD works in Pocketsphinx and if is possible to get
better performance,disabling it for example and using an external VAD.
I made several tests with Audacity to change silence period in my wav files
and I discovered that VAD is able to discover utterance even if there isn't a
silence period before the utterance.
Anyway the audio file must be long at least 3.136 seconds.
Any suggestion would be appreciated
Thanks in advance
Regards
Marco
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi everybody!
I 'd like understand how VAD works in Pocketsphinx and if is possible to get
better performance,disabling it for example and using an external VAD.
I made several tests with Audacity to change silence period in my wav files
and I discovered that VAD is able to discover utterance even if there isn't a
silence period before the utterance.
Anyway the audio file must be long at least 3.136 seconds.
Any suggestion would be appreciated
Thanks in advance
Regards
Marco
Is there anybody that can give me an hint about Voice Activity Detection works
on Pocketsphinx?
It's important for me!
Marco