Perhaps realtime API diarization is coming?

[multitechvisions]

Was looking through the realtime API documentation, and came across the following:

• https://platform.openai.com/docs/api-reference/realtime-server-events/conversation/item/input_audio_transcription/segment

Speaker? Could this be a leak, giving us a sneak peak at a new diarization capability?

[mferna23]

Have you tested if it is working?

[multitechvisions]

I got an error when I included it in the object, so not usable yet.

[multitechvisions]

https://platform.openai.com/docs/models/gpt-4o-transcribe-diarize