search

MahmoudAshraf97 / whisper-diarization

Automatic Speech Recognition with Speaker Diarization based on OpenAI Whisper
BSD 2-Clause "Simplified" License
2.44k stars 238 forks source link

word_timestamps - IndexError: list index out of range #178

Closed Reinmor closed 1 week ago

Reinmor commented 2 months ago

Hello.

Please help me with a few errors.

Processing a stereo call recording (.wav). One channel is clean, on the second channel the voice assistant speaks. I get the following error: 

  File "/home/ubuser/D/test5.py", line 542, in <module>
    word_timestamps = filter_missing_timestamps(
  File "/home/ubuser/D/test5.py", line 334, in filter_missing_timestamps
    if word_timestamps[0].get("start") is None:
IndexError: list index out of range

How can I fix it or is it better to analyse only mono recordings?

MahmoudAshraf97 commented 2 months ago

this has nothing to do with the channels as it's converted internally before processing, please upload the audio file to reproduce the issue