Title: Issues with Binary File Mapping of Zoom Recordings on Make.com
I am working on a scenario at Make.com where I need to automate the download and processing of Zoom meeting recordings. However, I am encountering a specific issue with the mapping of binary files that Zoom provides.
Objective: My goal is to automatically download Zoom recordings (specifically MP4/M4A files) and then convert them to a compatible audio format for further processing (such as transcription) with Whisper, for example.
Problem: When I download the recording file from Zoom via Make.com, the file is received in binary format. I am having difficulty mapping or converting this binary file to a usable format for additional steps in Make.com.
- Set up an event trigger on Make.com for when a Zoom recording is complete.
- Used a Zoom Cloud Downloading module to download the file, which I receive in binary format.
- Attempted to map this binary file to convert it to an audio format, but this is where I am stuck.
Question: Has anyone faced a similar issue or have any ideas on how I can properly convert or map these binary files for further processing on Make.com?
I would appreciate any guidance or advice you can provide.
Thank you in advance,
I’ll answer in English as my Spanish is really bad - I hope it works!
I think you want to transcribe a Zoom recording and then use OpenAI to do anything with this text. You’ll need to use a tool to convert video → MP3 and then use this within the OpenAI Module.
You could use something like: MP4 to MP3 | CloudConvert.
I developed my own solution using custom code but staying true to No-Code CloudConvert is your way to go.
You might want to check if mp3 is the best format. As long zoom meetings might create MP3 files which are still too big (+25MB).
Hope it helps!
The thing is that I am not capable of downloading the MP3 nor the MP4. I got this output from the Zoom module:
Which is a binary file that I cannot map into OpenAI Whisper module. Maybe your solution still works, but I am not familiar to CloudConvert and as far as I see it only converts MP3/4.
Thanks for your fast respond,
“Edited”. Thanks for pointing it out.
Just in case, I am directly downloading de M4A, not MP4. So it should be the audio file.
Hi @Jordi-IA, if you have the audio that’s even better than video! Don’t you simply need to map the data into the OpenAI Module into the file-data field? What’s the error you receive?
YES! Exactly, but as I stated in my initial Q, when I map to OpenAI module, it’s immediately unselected. The changes are not saved. This is the problem I am encountering.
When I re-open:
well, it’s still mapped.
Within File it chooses “Zoom…” because you don’t change the settings from the default.
If you e.g. change the filename, then it stays in this “mapped” settings.
So the problem now is finding why although we are mapping a file, it is still treated as a binary file.
I have been doing some trials uploading the file into Google Drive, but it is not recognised.
Thank you for your help Richard,
This isn’t as simple as needing to adding the file extension so that google recognises what to do with it is it?
I may just be totally wrong but I’ve had issues like that before.
Sorry for the ignorance Simon, but where can I add the file extension to Google? It seems that Google Drive is not prepared to upload audio/video files. The extensions that appear are mainly DOC, CSV, …
tbh I was talking generally rather than specifically about the GoogleDrive component as I’ve not used it, but I would assume you would just add it to the filename in the upload file component.
So just adding .m4a after the filename you’ve mapped into the field.
Nice catch there. So, motivated to your proposal I tried it and it works well, but I still think it is very inefficient to upload/download a file when I am already downloading it from Zoom.
Thank you very much Simon, will keep trying it in order to see if I can skip this step.
And if you adjust the filename within the Whisper Module directly?
- Filename (zoom).m4a
- Filedata the data you get from Zoom
It is what I was trying, and it works well!!
Thank you all Make community. Hope this Issue helps other people.