Integrating Audio Diarization Module into RescueBox

[nmj7]

The dependencies are;
1.onnxruntime==1.21.0
2.git+https://github.com/openai/whisper.git
3.ffmpeg-python
4.flask-ml==0.2.5
5.pyannote.core==5.0.0
6.pyannote.audio==3.3.2
7.pyannote.database==5.1.3
8.pyannote.metrics==3.2.1
9.pyannote.pipeline==3.0.1

Models:
1.pyannote/speaker-diarization-3.0
License : MIT License
2.whisper-medium.en
License : MIT License

[prasannals]

Thanks for the PR.

You should be removing flask-ml as a dependency (I remember you telling me that you're working on this).

Is this the license for pyannote/speaker-diarization-3.0? https://huggingface.co/pyannote/speaker-diarization-3.0/blob/main/LICENSE

If so, that is the MIT license and it that'll be good to use with RescueBox.

Everything else looks good in the initial review!

[nb950]

merged to https://github.com/UMass-Rescue/RescueBox/tree/plugins_from_spring_2025

[nb950]

merged to https://github.com/UMass-Rescue/RescueBox/tree/plugins_from_spring_2025