Create Transcription

I’m trying to use the “Create Transcription” AI block, and passing it a .webm file link from the Audio Recorder custom component in UI Builder. I receive this error, but I’m unsure how to change the configuration of the “Create Transcription” AI block to select a different response_format.

HTTP status code: 400. Body: { “error”: { “message”: “response_format ‘verbose_json’ is not compatible with model ‘gpt-4o-mini-transcribe-api-ev3’. Use ‘json’ or ‘text’ instead.”, “type”: “invalid_request_error”, “param”: “response_format”, “code”: “unsupported_value” } }

My overall goal is to enable voice control / dictation for my app. Thanks!

Hello @Laura_Kennedy

Welcome to our community and thank you for trying out Backendless.

I’ll take a look now.

Regards,
Volodymyr

I was able to reproduce the problem.
I have created internal ticked, and we will notify you when it will be solved.

Regards,
Volodymyr

Thank you!!

Hi @Laura_Kennedy ,

We have released fix for your issue. Could you please try and confirm that error is gone?

Regards, Andriy

1 Like

Yes! It works great now! Thank you!

1 Like