I am using whisper to transcribe a Google Drive File object, not a URL. The file is a 6.3 mb MP3 that is 26 minutes long, mono audio with a sample rate of 22khz. It triggers the following error in the Whisper stage of the Zap.
504 Gateway Time-out.
Failed to create a transcription in OpenAI (GPT-3, DALL-E, Whisper)
<html> <head><title>504 Gateway Time-out</title></head> <body> <center><h1>504 Gateway Time-out</h1></center> <hr><center>nginx</center> </body> </html> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page --> <!-- a padding to disable MSIE and Chrome friendly error page -->
The same Zap works fine when tested with another file that is 10 minutes long and 2.5mb.
OpenAI state that the Whisper API has a file limit of 25mb so my 6mb file should parse fine.
I found a similar issue with a user using the OpenAI API with Whisper “Whisper API server error for long (not big) files”: https://community.openai.com/t/whisper-api-server-error-for-long-not-big-files/135563
I believe this issue I am seeing is due to a bug in either the Zapier implementation of Whisper, or the Whisper API, as it should support files up to 25mb.
Is anybody else having a similar issue and are you perhaps able to provide support on this?
Many thanks