I signed up for PipeDream, and everything was working smoothly. I was able to get audio files turned into notes in Notion - then I started getting an error saying “Auto Disabled
This workflow was disabled because it used features that were not available on your billing plan.”
Turns out, PipeDream enrolls users to a free trial of the Advanced plan on signup. The Thomas Frank code enabled auto-retry errors in workflow settings, which was an Advanced feature, and not available on the Basic plan. I disabled auto-retry on the workflow, and then it worked…once. After that, I received this message, and PipeDream’s customer support indicated the error is being raised from OpenAI but it looks like it’s being handled by a custom part of Thomas’ code.
Any help is appreciated!
Error
An error occured while attempting to split the file into chunks, or while sending the chunks to OpenAI. Full error from OpenAI: 400 Audio file is too short. Minimum audio length is 0.1 seconds.
DETAILS
at Object.chunkFileAndTranscribe (file:///tmp/pdg/dist/code/1f555395881e5b79fbe9431588e45c7f9763b0394fdc7e0eaabc4ff53300ac5d/code/Notion-Voice-Notes.mjs:406:11)
at process.processTicksAndRejections (node:internal/process/task_queues:95:5)
at Object.run (file:///tmp/pdg/dist/code/1f555395881e5b79fbe9431588e45c7f9763b0394fdc7e0eaabc4ff53300ac5d/code/Notion-Voice-Notes.mjs:1990:22)
at null.executeComponent (/var/task/launch_worker.js:294:22)
at MessagePort.messageHandler (/var/task/launch_worker.js:791:28)
Based on the error message you’ve shared, it seems the issue arises from the audio file being too short. The OpenAI API is returning a 400 error with the message “Audio file is too short. Minimum audio length is 0.1 seconds.” This indicates that the audio file you’re trying to transcribe does not meet the minimum length requirement set by OpenAI for transcription.
To address this issue, you should ensure that the audio files you’re attempting to transcribe meet the minimum length requirement of 0.1 seconds. If you’re programmatically generating or manipulating audio files before sending them to OpenAI for transcription, verify that these operations do not result in files shorter than the minimum length.
If you’re confident that the audio files meet the minimum length requirement and you’re still encountering this error, it might be worth examining the part of the code responsible for splitting the file into chunks (as mentioned in the error stack trace). There could be an issue in how the audio file is processed or split before being sent to OpenAI.
Since the error mentions a custom part of Thomas’ code, specifically at chunkFileAndTranscribe function, reviewing this function for how it handles audio files could provide insights. Unfortunately, without access to the specific code, I can’t provide a detailed solution. However, ensuring audio file length and reviewing the file processing logic would be good starting points.
For further assistance with the code or if you encounter other issues related to Pipedream’s billing plans or features, I recommend reaching out to Pipedream’s support team directly at Support - Pipedream. They can provide more specific guidance based on your account details and the features you’re trying to use.
I’m a bot powered by Pipedream and GPT-4. I’m still learning, so please double-check my answers and code! Learn how to build your own.
