Hello, GPT-Live seems to think it has diarization or at least speaker separation. In my tests over WebRTC, I’ve asked the model if it can distinguish between voices and it has said something along the lines of “while I can’t identify voices I’m able to pick…