News for hackers. Stuff that matters.

A comment on Insanely Fast Whisper: Transcribe 300 minutes of audio in less than 98 seconds. Parent comment

Re:Insanely Fast Whisper: Transcribe 300 minutes of audio in... by siraben on Tuesday November 14, 2023 @01:19PM (#38267219)
Is diarization only possible with stereo audio at the moment in whisper? If the voices aren't left/right split I don't think one can get them separated yet.
[ Parent | on HN ]
Re:Insanely Fast Whisper: Transcribe 300 minutes of audio in... by copypirate on Tuesday November 14, 2023 @01:25PM (#38267316)
https://github.com/Wordcab/wordcab-transcribe

Author of Wordcab-Transcribe here. We use faster-whisper + NeMo for diarization, if you want to take a look.

[ Parent | on HN ]
Re:Insanely Fast Whisper: Transcribe 300 minutes of audio in... by pen2l on Tuesday November 14, 2023 @01:22PM (#38267272)
Whisper cannot perform speaker diarization at the moment, WhisperX and other solutions that can tend to use pyannot https://github.com/pyannote/pyannote-audio
[ Parent | on HN ]
Re:Insanely Fast Whisper: Transcribe 300 minutes of audio in... by theossuary on Tuesday November 14, 2023 @02:20PM (#38268155)
Nvidia's Nemo has some support for speaker recognition and speaker diarization too.

https://docs.nvidia.com/deeplearning/nemo/user-guide/docs/en...

[ Parent | on HN ]