Faster-Whisper Support?

[JvilleDev]

Can u add Faster-whisper support?
Or How can i do that?

[NosgothicRealm]

Hello!
I tried to implement this app using WhisperX because of the diarization option and the
faster_whisper's models.

But I have problems parsing the audio to Whisperx
This app 'normalizes' the audio and that works for openai's whisper, but not for Whisperx (I'm no expert in audio format)

If at some point I can implement this app using faster_whisper/whisperx I'll let you know.

sorry for my English!

[lukkzien]

Hello! I tried to implement this app using WhisperX because of the diarization option and the faster_whisper's models.

But I have problems parsing the audio to Whisperx This app 'normalizes' the audio and that works for openai's whisper, but not for Whisperx (I'm no expert in audio format)

If at some point I can implement this app using faster_whisper/whisperx I'll let you know.

sorry for my English!

Hello, were you able to accomplish it?

[NosgothicRealm]

But I have problems parsing the audio to Whisperx This app 'normalizes' the audio and that works for openai's whisper, but not for Whisperx (I'm no expert in audio format)

If at some point I can implement this app using faster_whisper/whisperx I'll let you know.

Until some point, yes, but I'm still having issues because there are still problems when audio is shared using 'normalization'
I'm trying to implement something similar, but using WhisperX because that repo uses the diarization too.

If I have something good for real-time I'll share it :)

[HardikJain02]

Have you got anything for real-time?