[FEATURE REQ] Support for new transcription model, gpt-4o-transcribe-diarize

[Hobbeslionheart]

Describe the feature or improvement you are requesting

OpenAI released a new model for transcription, gpt-4o-transcribe-diarize. This most likely requires modifying existing fields and adding some new fields to AudioTranscriptionOptions and AudioTranscriptionFormat to support this model.

Model Page
Speaker Diarization section
API for transcription

Additional context

No response

[jsquire]

Hi @Hobbeslionheart. Thanks for reaching out and for your inquiry. This model is a recent addition and is not yet represented in the REST API spec used for client generation. Once the REST API spec is updated to include this area and a new version published, we'll include it in the client.

In the meantime, you can access the necessary additional properties by using the protocol overloads or the new JSON Patch extensions included with v2.6.0 once that is published. (an example of the syntax is available in the Example07_InputAdditionalProperties sample)