async-openai
async-openai copied to clipboard
AudioTranscription has non-existing enabled field
For the realtime types, AudioTranscription currently has an enabled field, but the API does not have this. This triggers an error when trying to set an audio transcription on session update:
called `Result::unwrap()` on an `Err` value: invalid_request_error [code=Some("unknown_parameter"), param=Some("session.input_audio_transcription.enabled")]: Unknown parameter: 'session.input_audio_transcription.enabled'.
This is addressed in https://github.com/64bit/async-openai/pull/391
Yeah I also opened a PR to fix but it never got attention. Not sure if this is a priority.