i’m trying to add audio files to my dropbox and then have openai whisper transcribe the audio to that dropbox folder so that I can read the .txt file. I’m getting this error:
The operation failed with an error. [413] Maximum content size limit exceeded. Googleing I see that I probably need to chunk the audio files down to max size pieces of 25mb.
How can I do that in make?
2 Likes
Looks like this has been discussed before but there wasn’t any resolution Onedrive download audio file chunking route to OpenAI API
Anyway I suggest looking into Google Cloud Speech, which allows you to process way larger files.
For more information on this, see
2 Likes
You can use file compression to lower the size of your file.
I used CloudConvert and used this setup:
I was able to lower the size of my file from 40MB mp3 to 8MB M4A. M4A format is also accepted by Whisper.
Also, you can do this using CloudConvert free account
6 Likes