Hello, I would like that when using Whisper I could separate the transcriptions into channels, since when transcribing a phone call I need to know what person 1 said and what person 2 said, with Whisper can I do that?
Hi Juan
This is called diarisation. There are various libraries for that out there that you can use with Whisper.
Nope that wont work. Splitting channels and transcribed in not practically a good solution. See this discussion 1026 in Github Whisper.
Whisper should have this and it is very much required. Google speech to text having this feature.