Troubleshooting Audio Input Issues with Lindy and Telegram Bot | Lindy Community

Lindy Community Icon

Troubleshooting Audio Input Issues with Lindy and Telegram Bot | Lindy Community

Deuce T.
·
·
can you use the call agent and just call lindy support bot?
Nitai B.
·
·
Try having a Trigger based on a telegram message received, maybe it will have the audio file payload that you can then use "Download File" which is a Lindy Attachment Action.
Loading attachments
👀1
Jake
·
·
Unless I am confused, I have deff used the Lindy support bot. That's how I know that I need to "GET" the url of the voice message first and that's where things go wrong, no matter what I do it fails, even though I send a voice message Lindy seems to refuse to take any input from Telegram about anything other than text messages, it just seems to ignore the existance of other fields.
Jake
·
·
So the issue is that Telegram should have a field called "message.voice.file_id" which contains the voice messages ID, which I need to run a GET http request at to get the URL for the voice message; however, that option doesn't appear as an output from the telegram message trigger and when I enter it manually it doesn't work... It's as if Lindy is refusing to acknowlege it's existance.
Jake
·
·
Is it possible this is a Lindy limitation, that when it says "telegram message received" and "observe telegram messages" it literally means ONLY text messages and not voice messages? If so, that is probably enough to send me back to n8n..
Jake
·
·
Okay, so through connecting Telegram via webhook rather than the actual Telegram trigger I now have the JSON with the voice message file ID but it seems impossible to get it out of there. Is there not a simple step for extracting text from a JSON output and selecting which bit you want to keep? I have tried all sorts of code and set variable options and non have worked..
Alon J.
·
·
Hey Jake! Are you able to send over the task URL? What is happening when you use the download action? How to share your Lindy task URL: 1. Click your Lindy agent → click “tasks” 2. Find the task you need help with 3. Click the specific task (you should see the full execution) 4. Copy that page URL from your browser 5. Paste the URL in your support request
Jake
·
·
https://chat.lindy.ai/clearlight/lindy/telegram-audio-transcription-agent-68a77f1e5[…]&testRunId=68a85c2ef05b61ecb2f6f97e&e=68a859aafd8a620c831076ed
Jake
·
·
but was this question in response to my more recent support request that I posted about, rather than this thread? https://lindycommunity.slack.com/archives/C07FX8W5ME3/p1755865920719179
Jake
·
·
now I am trying it with Gladia, without downloading the file. and Gladia is giving me the same error "the audio file or url field is requiered" but directly above the rror, it's showing the URL field, filled out, with a URL, that if I copy and paste into my browser downloads the file. https://chat.lindy.ai/clearlight/lindy/telegram-audio-transcription-agent-68a77f1e5[…]&testRunId=68a85c2ef05b61ecb2f6f97e&e=68a86214fbbe3683ca2ec395
Alon J.
·
·
Hmmm yeah this looks like a bug with the Gladia integration 😕
Alon J.
·
·
I'm reporting to my eng team!
Alon J.
·
·
I'm getting the same error
Jake
·
·
So Gladia, Deepgram, and the OpenAI transcription things all don't work?
Jake
·
·
Is there any way currently working on lindy to transcribe audio/voice messages atm?