Whisper & python - set more parameters?

API
1

Reading through the Whisper Quickstart Guide it explains how to simply request a transcription: openai.Audio.transcribe("whisper-1", audio_file)

It proceeds to say that to add extra parameters, just add --form lines which I imagine refers to the REST API, is there a way to set these parameters such as response_format in python or can this only be done with a CURL request?

2

Hello, I have the same issue. Have you managed to solve it? Thanks

3

I’ve managed to solve it. You just need to add the parameter at the end of the request:

transcript = openai.Audio.transcribe(“whisper-1”, audio_file, language=“es”)

or

transcript = openai.Audio.transcribe(“whisper-1”, audio_file, response_format=“text”)

1 Like