search

thomasmol / cog-whisper-diarization

Cog implementation of transcribing + diarization pipeline with Whisper & Pyannote
https://replicate.com/thomasmol/whisper-diarization
165 stars 51 forks source link

I did not get the complete transcript #11

Closed learner0333 closed 6 months ago

learner0333 commented 6 months ago

I am using the replicate for the transcript and Diarization. I got very good results. But I am not getting the full transcript and diarization data for the following video:

https://www.youtube.com/watch?v=FLEaCRvGmKM

I have downloaded the video and passed it to replicate. I get the transcript and diarizationf of first 18 minutes or so. will you please update me about it?

thomasmol commented 6 months ago

Hi! this might be due to the model thinking the audio has ended. This happens sometimes during long pauses where there is no speech (singing is also difficult to recognize). I don't have a good solution for you right now, the only thing I can suggest is try cutting the video into segments where there is only speech. Let me know if this helps!