Multilingual Telegram Voice & Text Bot with ElevenLabs TTS & LangChain

This n8n workflow powers a sophisticated Telegram bot capable of processing both voice messages and text inputs. For voice queries, it transcribes audio using ElevenLabs STT, analyzes with AI models like Groq or Gemini via LangChain agents, and responds with natural speech generated by ElevenLabs TTS. Text queries bypass audio processing for quick text replies, ensuring efficient interactions.

Key benefits include smart response logic that matches input type, multilingual support via Telegram's language detection, and conversation memory for context-aware chats. It supports extensible AI tools for tasks like API calls, database queries, or custom functions, enabling multi-step reasoning such as fetching crypto prices and analyzing trends.

Ideal use cases span voice-first customer support, crypto analytics assistants, multilingual FAQs, and educational tutors. Highly customizable, it requires Telegram Bot Token, ElevenLabs API key, and an AI provider key. Perfect for businesses seeking advanced, interactive AI assistants that save time on development and enhance user engagement.