search

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

kernel crashed at whisper transcription cell. #104

Closed manjunath7472 closed 9 months ago

manjunath7472 commented 9 months ago

Running locally on RTX 4000 card. kernel crashed at whisper transcription cell after 10 min of processing. Is there any time length limit for input audio file? Small length file had no issues say 1 min audio file.

manjunath7472 commented 9 months ago

In transcribe cell, we need to remove below lines to avoid kernel crash while transcribing. It also solves problem where all dialogues are stuffed under one speaker.

del whisper_model torch.cuda.empty_cache()