Audio file might be corrupted or unsupported

API Bugs
1

Hi everyone! :waving_hand:

I’m currently facing an issue with the new transcription capability of the OpenAI API using the gpt-4o-transcribe and gpt-4o-mini-transcribe models.

When I try to submit an audio file (in a supported format, as specified in the official documentation), I receive the following error response:

{
  "request_id": "req_1e3293a3f121081a41273fdc2394c0c2",
  "error": {
    "message": "Audio file might be corrupted or unsupported",
    "type": "invalid_request_error",
    "param": "file",
    "code": "invalid_value"
  }
}

However, I’ve confirmed that the file is not corrupted and is in a valid format (I’ve tested with .mp3, and .wav). The audio plays fine in various players, and I was also able to transcribe it successfully using whisper-1.

Is anyone else encountering this issue with the GPT-4o transcription models? Are there any undocumented limitations, or suggestions on how to structure the request differently?

Any insights or help would be much appreciated!

1 Like
2

Is it happeninng with all files or a particular one?

Here are a few related cases that may help:

1 Like