Error while converting speech to text - Jotforms, Https-Get-OpenAI Whisper

Hello All,

I am new to make and trying things out. I am using Jotforms to capture an audio record (.WAV file), passing that to HTTPS- Get module and then connecting it to OpenAI Whisper module to convert the voice to text. I am getting this error below " RuntimeError

[400] Invalid file format. Supported formats: [‘flac’, ‘m4a’, ‘mp3’, ‘mp4’, ‘mpeg’, ‘mpga’, ‘oga’, ‘ogg’, ‘wav’, ‘webm’]


OpenAI (ChatGPT, Whisper, DALL-E)

Automatic error handler

If you want to handle this error automatically, choose one of the following options. This will create a new error-handler route in your scenario. You can then expand the route in any way you like.

Ignore all errors
Ignore errors of type RuntimeError".

How do I resolve this?

Can you share the output bundle of the HTTP app and the configuration of your OpenAI module?

ALso move the clock icon to the Jotform module since currently it’s pointing at a null module. You can also just delete the grey + module by right clicking on it and the clock will move to Jotform automatically.

1 Like

Hi Alex, here is the output bundle: Somehow http-Get module shows the filename as .html.
get_output.txt (259.6 KB)

Below is the configuration for OpenAI:

I have tried OpenAI with “Get a file” option as well and this is the error I get:

Your file data parameter is set to filesize. It’s wrong though. That should be the the binary data buffer called filedata or something like that.

Also your http call is not returning the wav file but rather something else entirely. You need to retrieve the wav file from the submission so you need to call the direct download link for the file and right now you’re getting some sort of webpage from Jot form so that’s not right either.

Hi @Anir_Banerjee

As your file is .html, it should be in any audio supported format listed. However, whisper model doesn’t work fine with all audio formats. Hence, you will need to go with cloud convert to convert the file into .mp3 and then feed the file to GPT.

If you require additional assistance, please don’t hesitate to reach out to us.
MSquare Support | Book Live Implementation
Visit us here
Youtube Channel

1 Like