🚩 Report: Not working
Traceback (most recent call last):
File "/home/user/app/app.py", line 370, in transcribe_webui
result = self.transcribe_file(model, source.source_path, whisperLangCode, task, vadOptions, scaled_progress_listener, **decodeOptions)
File "/home/user/app/app.py", line 515, in transcribe_file
result = self.process_vad(audio_path, whisperCallable, self.vad_model, process_gaps, progressListener=progressListener)
File "/home/user/app/app.py", line 648, in process_vad
return vadModel.transcribe(audio_path, whisperCallable, vadConfig, progressListener=progressListener)
File "/home/user/app/src/vad.py", line 185, in transcribe
sub_task_total = 1/len(merged)
ZeroDivisionError: division by zero
Hi ednsinf,
Thank you for reporting this issue. Upon initial investigation, it seems that some audio files return null values when executing get_transcribe_timestamps, which causes a "division by zero" error in subsequent processes. However, what’s puzzling is that this issue only occurs on huggingface spaces and cannot be replicated when running the program locally...
Hi ednsinf,
The issue with whisper-webui-translate has been resolved. The root cause was related to the version of the Silero-VAD model used for speech detection.
https://github.com/snakers4/silero-vad/issues/515
Thank you!