Request on client side to gpt-4o-mini-transcribe model

API
1

This is my code on client side:

const formData = (blob: Blob) => {
  const body = new FormData();

  const file = new File([blob], "speech.mp3", { type: "audio/m4a" });

  body.append("file", file);

  body.append("model", "gpt-4o-mini-transcribe");

  body.append("language", "en");

  body.append("response_format", "json");

  return body;
};

const sendAudio = async ( ) => {
  try {
    const response = await fetch(
      "https://api.openai.com/v1/audio/transcriptions",
      {
        method: "POST",
        body: formData(blob),
        headers: {
          Authorization: `Bearer ...`,
        },
      },
    );

    const json = await response.json();

    return json
  } catch {
    return "";
  }
};

And getting this error

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

What am I doing wrong?

2

It worked with body.append("model", "whisper-1");