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

HF models #109

Closed peregilk closed 7 months ago

peregilk commented 9 months ago

How much work would it be to also get this working with the finetuned HF models?

MahmoudAshraf97 commented 9 months ago

Just convert your model to CT2 format as explained Here and pass the path to the model name argument

peregilk commented 7 months ago

Great. Thanks. I was able to run the models when generating and storing the CT2-files locally. I tried pushing the CT2-files to the hub, but I was unable to find a way to load them from the command line.