I want to develop a speech to text application with OpenAI. The use case is to create workflows on spoken text. I use svycaptureAudio to make a audio recording, and I sent this to OpenAI chat completions url.
When I sent this recording to OpenAI for transcription I get an error:
{
“message”: “Invalid value: ‘opus’. Supported values are: ‘wav’ and ‘mp3’.”,
“type”: “invalid_request_error”,
“param”: “messages[0].content[0].input_audio.format”,
“code”: “invalid_value”
}
Is there anyway to convert this opus format to wav of mp3?