Transcription with diarization using Whisper and pyannote.audio. Simplifies code and contains recommendations to fix dependency issues other repos face.
Other similar repos:
- https://github.com/m-bain/whisperX: Quite similar, but much simpler code here. whisperX also tries to show only one speaker per section which does not work that well sometimes.
- https://github.com/MahmoudAshraf97/whisper-diarization: Quite similar too. Recommended some fixes here that are unresolved in that repo last I checked.