Your API calls to OpenAI are outdated.
Use Responses API for transcription:
POST https://api.openai.com/v1/responses
With:
model: gpt-4o-transcribe or gpt-4o-mini-transcribe
audio file as input
This is the supported path going forward.
It seems that you are still calling https://api.openai.com/v1/audio/transcriptions which now fails.
Your API calls to OpenAI are outdated.
Use Responses API for transcription:
POST https://api.openai.com/v1/responses
With:
model: gpt-4o-transcribe or gpt-4o-mini-transcribe
audio file as input
This is the supported path going forward.
It seems that you are still calling https://api.openai.com/v1/audio/transcriptions which now fails.