Voice AI Telegram chatbot 100% self-hosted open source

Hello, n8n-AI-flow-people, magic people!
I’m just a beginner here, so will be happy if any help.

I’m running self-hosted n8n-ce, Ollama, open-webui and other homelab related services in docker.

n8n is a greate tool! I found some almost ready-to-use workflows (thank you all the people) and I was able to build local AI agents and Telegram bots.
BUT!
I do not have a Open AI api key, so all of my helpers are Voiceless.
Could you please advise the way to give them voice without using Whisper and any Open AI models.
Maybe there is a way to use some tool like LiteLLM, different TTS and STT ?

Thank you in advance!

  • n8n version: latest community
  • Database (default: SQLite): any - have bunch of them in docker
  • Running n8n via Docker
  • Operating system: linux-based

It looks like your topic is missing some important information. Could you provide the following if applicable.

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

Hello!

Info’s got added

I am using the following Python library to implement speech functionality:

However, I encountered the following issue. Can you help me?

Thank you!
I will try.
I’m not a programmer, but thanks AI - they help me with coding )