A Telegram bot that enables voice and text-based AI chat interactions. It's designed for users who want a conversational AI integrated into their Telegram chats, offering a seamless way to communicate with AI using voice or text and receive AI-generated responses in kind.

How It Works

This project leverages Cloudflare Workers to host a Telegram bot backend. It integrates with various AI models for speech-to-text (Whisper), text generation (Llama), and text-to-speech (MeloTTS, SiliconFlow, Deepgram). The architecture allows for flexible configuration of AI providers and models, enabling users to choose their preferred backend services for different functionalities.

Quick Start & Requirements

• Primary install/run command: Deploy using Cloudflare Workers via wrangler.
  1. pnpm install
  2. pnpm install -g wrangler
  3. Configure Cloudflare credentials (wrangler login or environment variables).
  4. Create a KV namespace: wrangler kv namespace create tgvoicechat.
  5. Set secrets: tg_token (required), siliconflow_token (optional).
  6. Configure tts_provider (workers, siliconflow, or deepgram).
  7. Deploy: pnpm run deploy.
  8. Initialize webhook: Visit https://<your-worker-domain>/init once.
• Prerequisites: Node.js with pnpm, Cloudflare account, Telegram Bot Token.
• Links:
  • Bot Demo: https://t.me/tg_voice_ai_bot
  • Cloudflare Workers: https://workers.cloudflare.com/

Highlighted Details

• Utilizes Cloudflare Workers for serverless deployment, potentially offering scalability and cost-efficiency.
• Supports multiple AI backends for TTS, including Cloudflare's own models, SiliconFlow, and Deepgram.
• Employs specific AI models: Whisper v3-turbo, Llama 4-scout-17b-16e-instruct, and various TTS models.

Maintenance & Community

No specific information regarding maintainers, community channels (like Discord/Slack), or project roadmap is available in the provided README.

Licensing & Compatibility

The license is not specified in the README. This lack of explicit licensing information may pose compatibility concerns for commercial use or integration into closed-source projects.

Limitations & Caveats

The project's setup requires familiarity with Cloudflare Workers and wrangler. The absence of a specified license is a significant caveat for adoption. Configuration details for TTS providers and models are spread across the README and wrangler.json, requiring careful review.