Can't do Telegram speech to text with Gemini

Questions
1

Hey, I’m trying to transform an audio file from telegram into a text by using Gemini. But I see that almost everyone is using the OpenAi node to do it.
ANy help or suggestion would be appreciated thanks.

Describe the problem/error/question

Since there isn’t a specific node to do this with Gemini, I tried to call the API directly with a HTTP request but I can’t make it work. It maybe is because of the format of the file but I’m not sure.

What is the error message (if any)?

Bad request - please check your parameters

  • GenerateContentRequest.contents: contents is not specified

Please share your workflow

Share the output returned by the last node

The output is the error

Information on your n8n setup

  • n8n version: 1.84.1
  • Database (default: SQLite): SQLite
  • n8n EXECUTIONS_PROCESS setting (default: own, main): own, main
  • Running n8n via (Docker, npm, n8n cloud, desktop app): 8n cloud
  • Operating system: W11
2

The Gemini API requires a few more steps before transcribing the audio.
I was able to make it work like this. Try it and let me know if this solves your problem.

.

By the way, I see that you still don’t have a Google Gemini credential configured. Take a look at how I use the credential on my HTTP node and try to do the same. It’s safer to use credentials than it is to write the key in the workflow.

3

Your simply the goat man thank you very much

4

Thank you so much! This is absolutely incredible. I have been looking for an open AI alternative and it is an absolute god send to find this here.

The n8n comunity is great!