I’m new here and I hope I understood the rules here correctly, if not please let me know and forgive me!
But I’m not here just to introduce myself (because that not what that section is for), but to ask you for a bit help and advice. I found on YouTube a tutorial by Thomas Frank about making perfect notes by recording them from your phone (How I Use AI to take perfect notes...without typing - YouTube). Sounds ideal, especially for a person with disability like me, but I’m stuck when I try to create a transcription on OpenAI,(How I Use AI to take perfect notes...without typing - YouTube - in this moment of the video) because of an error:
Error
/tmp/Test Audio.mp3 does not exist
DETAILS
at Object.run (file:///tmp/__pdg__/dist/code/8a146976f5aa39e7dc8c29ab343b525d8f1fb0f1a9d7c29c9eb82d2c59bf2a83/code/actions/create-transcription/create-transcription.mjs:203:15)
at global.executeComponent (/var/task/launch_worker.js:139:53)
at MessagePort.messageHandler (/var/task/launch_worker.js:598:28)
Steps before have been tested and successful, so I’m sure what mistake I’m making here. Please, let me know and I very much appreciate your help and support!
Thanks a lot for your response @thomasfrank. I very much appreciate that!
Unfortunately I still didn’t have a chance to check that, because I was a bit busy. but I’ve got some time tomorrow schedule for that, so then I’ll let you know guys how successful I was.
You’re welcome @chickanan I hope you’ll find a way to make it work!
@chickanan this is likely due to one of your steps having the wrong name. Have you made sure all your steps are named exactly as shown in the tutorial?
You were right. I made an error with the naming since I used google drive instead of dropbox and I had to sort things out. But I did and it works like clockwork.
I found a really interesting usecase for it.
I am a foreign exchange student in Russia. So my lectures are in Russian and since it is not my first language, it was really difficult to take good notes.
Now, I record it and the output is in English by default on notion. Furthermore, I use notions translate feature and make a translated copy for my Chinese class mates too.
This system has the potential to make learning so much easier for foreign students.
I’m using the code version as I’m using Dropbox as my repository. The code works like a dream (no pun intended)… Except my transcription changes languages! Everything I record is in English but all of the titles except one transcribed in Spanish. And when I look in the one with the English title, the summary starts in English but ends in Spanish! Any ideas?
The text is generated by Open AI, so you can try to check your Open AI step in your Pipedream workflow if anything in the prompt mentioning Spanish. If not, I think that’s the issue on Open AI (instead of Pipedream). You can try to contact them via email, giving them the prompt you added and the transcript, so maybe they can fix it
Hi @thomasfrank : my configure failed at step of formatter:
all other above steps were tested successful, checked my coding also is the same as yours. include top and bottom lines:
export default defineComponent({
async run({ steps, $ }) {
},
})
not sure where went wrong.
@thomasfrank
This automation has been a huge game changer for me. I have been able to use this to journal consistently while I drive. It has allowed me to reflective for the first time.
BUT the workflow has stopped working. I have changed nothing about the workflow. It not longer triggers. I upload a the audio file to the google drive and look at pipedream and there is nothing. No event, no error messages. I checked the workflow again and it says it is functioning correctly. Do you have any suggestions on what is wrong?
@zachary.cook92 the workflow is not triggering at all? I’m still using it without issue, although I use the Dropbox version. I recorded multiple notes today successfully.
Perhaps the Google Drive trigger isn’t working? If so, that’d be an issue for the Pipedream team to address, as the trigger is in the public repository and not maintained by me.
@thomasfrank, I am getting an error at the last step of configuring notion_voice_notes. I have tried the google_drive_download step over and over again but still keep getting errors. The last error I received was:
Error
An error occurred while performing a moderation check on the transcript: An error occurred while performing a moderation check on chunk 5. The content of this chunk is as follows: And that makes it really good for learning how to work with API’s. And for example, if you wanted to build your own pokedex, you don’t have to go manually get all the data for it. You can use poke API to get all of the names, the movesets, the pictures, the sprites, they have it all. So it is basically set up to give you all this interesting data that you probably are at least somewhat familiar with to build a great example project. Error message: 429 Rate limit reached for text-moderation-007 in organization org-yv49riEAHDeWtbx7Jin8A5dG on requests per min (RPM): Limit 3, Used 3, Requested 1. Please try again in 20s. Visit https://platform.openai.com/account/rate-limits to learn more. Note that you can set Enable Advanced Settings to True, and then set Disable Moderation Check to True, to skip the moderation check. This will speed up the workflow run, but it will also increase the risk of inappropriate content being sent to ChatGPT. Note that you can set Enable Advanced Settings to True, and then set Disable Moderation Check to True, to skip the moderation check. This will speed up the workflow run, but it will also increase the risk of inappropriate content being sent to ChatGPT.
I waited more than 20s and tried again and got this error:
Error
An error occurred while processing the audio file: Failed to read audio file metadata. The file format might be unsupported or corrupted, or the file might no longer exist at the specified file path (which is in temp storage). If you are using the Google Drive or OneDrive versions of this workflow and are currently setting it up, please try testing your ‘download’ step again in order to re-download the file into temp storage. Then test this step again. Learn more here: How to Take Perfect Notes with Your Voice Using ChatGPT and Notion
I have cycled through these two errors for more than 20 minutes. I am not sure what I am doing wrong. I really do not want to set Enable Advanced Settings to True and then set Disable Moderation Check to True to skip the moderation check.
@ppatriq I’m a bit confused - your transcript is not passing moderation, but you don’t want to disable the moderation check?
I would recommend disabling it; in fact, new versions of the workflow do so by default. Moderation check was a good idea when I built this, but it’s proven to be too sensitive to be useful.
Hello, I am having issues in the same step. The previous step “Trigger” works fine and should bring what is inside “steps” forward for the Microsoft One Drive download step, but everything is alright above, but I am always getting this message.
Error
Oops, this workflow won’t work if you use the Generate Test Event button in the Trigger step. Please upload an audio file (mp3 or m4a) to OneDrive, select it from the Select Event dropdown beneath that button, then hit Test again on the Trigger step.
DETAILS
at Object.run (file:///tmp/__pdg__/dist/code/c73cc97bafd30cc8535633caa613654c37f82890c5175fdbcf949db3ed690676/component.mjs:37:11)
at null.executeComponent (/var/task/launch_worker.js:286:53)
at MessagePort.messageHandler (/var/task/launch_worker.js:785:28)
