search

MahmoudAshraf97 / whisper-diarization

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper
BSD 2-Clause "Simplified" License
3.28k stars 272 forks source link

Vocal separation #25

Closed PabloSergi closed 1 year ago

PabloSergi commented 1 year ago

Hello guys, can you advise a good ool or neural network just to separate the vocals from the rest of the noise?

MahmoudAshraf97 commented 1 year ago

Demucs is the best model I saw, it's used in this repo

PabloSergi commented 1 year ago

Demucs is the best model I saw, it's used in this repo

thx, looks like they stoped renew it too.. i used some time ago spleeter, same situation, last update 2 years ago)

MahmoudAshraf97 commented 1 year ago

I don't think that'll be a problem as long as it's performing well

mrgalindo commented 1 year ago

I always get "Source splitting failed, using original audio file." Any ideas on how to troubleshoot this?

MahmoudAshraf97 commented 1 year ago

@mrgalindo try to use demucs alone to get a more descriptive error message