Rizki
I know that speaker recognition is in development within Hedy on board whisper model. Therefore I am using Deepgram API to help me get this diarization function. The Deepgram API integration is running smoothly on my intel Mac Mini 2012 running windows 10. Nevertheless I cannot see the diarization result in Hedy.
To support my meeting workflow, could you pelase add two updates to the backend API configuration:
  1. Enable Speaker Diarization: Please add the diarize=true parameter to the Deepgram API call so the transcript accurately separates different speakers.
  2. Add Flux Model: Please add "flux" to the model selection menu alongside nova and base so I can select it when needed.
Best,
RSS