Same words repeated for entire transcript.

MahmoudAshraf97 / whisper-diarization

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper

BSD 2-Clause "Simplified" License

2.53k stars 243 forks source link

Open manjunath7472 opened 9 months ago

manjunath7472 commented 9 months ago

I added in transcribe. vad_parameters=dict(threshold=0.4, max_speech_duration_s=15)

Then it worked.

Is there anything else we can do??

manjunath7472 commented 9 months ago

Below is audio file to reproduce the issue. Audio.