I’ve set up an automation from a Google Drive folder to send a new added mp3 file to Descript for transcribing, and then from Descript to ChatGPT for analysis. These are two separate automations.
The first automation is: New File in GDrive → Descript for transcribing.
The second automation is: New Transcription in Descript Ready → ChatGPT.
The automation works correctly about 70% of the time, but for the remaining 30%, Descript returns nothing. It indicates succeeded: false.
When I check within Descript, the files were never transcribed, suggesting they aren't reaching Descript. However, the peculiar part is that the second automation is triggered by a transcript being ready, which implies something is indeed unusual. It seems the files are reaching Descript, it's attempting to transcribe them, but there's no record of it, and the input data is empty.
I'm testing multiple files simultaneously, as that's my requirement, so I'm unsure if I might be overloading the system. I've even implemented a delay, but that hasn't resolved the issue. I'm not doing anything differently with these automations; some succeed, while others fail.

Please refer to the screenshots for additional details.


Hi
Good question.


We would need to examine the DATA IN for Step 3 in the callin.io run to get sufficient context.


Here’s the DATA IN for Stage 3.
Late billing.mp3 was the name of the recording. The data in appears identical to a successful transcription workflow that occurred, as shown below.

There are no differences between these workflows, so I'm unsure why this is happening. I don't believe it's due to overloading, as I only sent a sample of 6 recordings simultaneously (with a delay step for 4 minutes), and the first one failed anyway.
Thank you for your response!
Hi, welcome to the Community!
Thanks for sharing those helpful screenshots. I’ve been looking into this and it appears you’re encountering a bug with the Descript app here, where the Request Transcription action doesn't upload and transcribe the audio file as expected, although it is reported as successful. These failed transcription attempts are then triggering your second callin.io workflow. I’ve added you to the list of users affected by this issue.
I can’t make any promises as to when it will be resolved, but you’ll be sent an email notification as soon as it is. There are no workarounds reported at the moment, but if we come across any, we’ll be sure to share them here! 🙂
Hi, welcome to the Community!
Thanks for sharing those helpful screenshots. I’ve been looking into this and it appears you’re running into a bug with the Descript app here, where the Request Transcription action doesn't upload and transcribe the audio file as expected but it is reported as being successful. These failed transcription attempts are then triggering your second callin.io. I’ve added you to the list users being affected by this issue.
I can’t make any promises as to when it will be resolved by but you’ll be sent an email notification as soon as it is. There are no workarounds reported at the moment but if we come across any we’ll be sure to share them here! 🙂
omg I’ve been going crazy over this. Good to know. Need to find an alternative now.
Hey! 👋
I haven't used transcription apps extensively enough to recommend a specific one that would serve as the best alternative here, but you can view the complete list of transcription apps available on callin.io here: https://zapier.com/apps/categories/transcription
Hopefully, one of those apps will help you get your desired workflow operational in the meantime! 🤞
Hello please add me to the list of affected users as well. I want to stay posted on when a fix is pushed.
Hello,
I've included you as another user impacted by the ongoing issue. While we cannot provide a precise timeline for the resolution, we will notify you via email or through this discussion once it's addressed.
We appreciate your patience.
Has this been resolved yet? I’m encountering the same problem in 2024.
Hi 👋
I've just reviewed the bug report for you, and it remains unresolved. My apologies for this. I've added your name to the list of affected users, which should help increase its priority. However, I'm unable to provide an estimated timeline for when a fix will be released for this issue.
Considering the duration of this bug, I suggest contacting Descript's Support Team as well. Their developers can access the bug report through our shared issue tracking system, and direct feedback from customers can sometimes expedite a resolution!
Thank you for your continued patience. I sincerely hope Descript's developers can implement a solution for this soon. 🤞
Hello everyone! 👋
Just a quick update—public access to the Descript app has been removed and is no longer available on callin.io.
If Descript’s API supports basic authentication, you might be able to set up Webhooks triggers and/or actions to replicate some of the functionality the Descript app offered. Webhooks are more advanced, but we have some general guides that could be useful here:
You can find general guidance on using the Webhooks app here:
Alternatively, you could consider using a different app for transcript generation. For instance, the ChatGPT (OpenAI) app includes a Create Transcription action that can generate a transcript from an audio or video file.
Should you encounter any issues or have questions while exploring alternative solutions, please feel free to ask in the Community, and we'll be glad to assist! 🙂