Closed MadhavVij closed 4 years ago
I suppose the problem that you have is Speaker diarisation and Dejavu is not intended to do so. This library detects the "exact" audio signal. There are some other libraries that could help you with the Speaker diarisation task, like pyAudioAnalysis.
Thanks a lot for your response.
That would be a new project idea. And a nice one.
I have tried speech recognition with this library and i got lot of noises, false detection. works perfectly for Music.
This library is not meant for that purpose - please read the README. Thanks!
My problem is to identify different people and recognizing their voice. Can I use this library and tweak things up to recognizing them (even if they say different things)?