TRTC AI TTS Service

A minimal Node.js server that provides AI conversation capabilities with Text-to-Speech (TTS) support for Tencent Real-Time Communication (TRTC).

Features

• 🎙️ Real-time AI conversation with voice support
• 🤖 Integrated LLM for intelligent responses
• 🔊 Text-to-Speech using MiniMax TTS
• 🌐 Web-based conversation interface
• 🚀 Simple single-file backend architecture

Quick Start

Prerequisites

• Node.js (v14.0.0 or higher)
• Tencent Cloud account with TRTC enabled
• LLM API access (OpenAI compatible)
• MiniMax TTS API access

Installation

1. Clone the repository:

git clone https://github.com/yourusername/trtc-ai-tts.git
cd trtc-ai-tts

2. Install dependencies:

npm install

3. Configure environment variables:

cp env.example .env
# Edit .env with your credentials

4. Start the server:

npm start

5. Open your browser and navigate to:

http://localhost:3000/conversation.html

Configuration

Create a .env file with the following variables:

# Tencent Cloud API
TENCENT_SECRET_ID=your_secret_id
TENCENT_SECRET_KEY=your_secret_key
TENCENT_REGION=ap-guangzhou
TENCENT_ENDPOINT=trtc.tencentcloudapi.com

# TRTC Configuration
TRTC_SDK_APP_ID=your_sdk_app_id
TRTC_SECRET_KEY=your_trtc_secret_key

# LLM Configuration
LLM_API_URL=your_llm_api_url
LLM_API_KEY=your_llm_api_key
LLM_MODEL=your_model_name

# MiniMax TTS
MINIMAX_TTS_GROUP_ID=your_group_id
MINIMAX_TTS_API_KEY=your_api_key
MINIMAX_TTS_VOICE_TYPE=your_voice_type
MINIMAX_TTS_API_URL=http://api.minimax.chat/v1/t2a_v2
MINIMAX_TTS_MODEL=speech-01-turbo

API Endpoints

Method	Endpoint	Description
POST	/conversations	Start a new AI conversation
DELETE	/conversations	Stop an active conversation
POST	/credentials	Generate TRTC user credentials

Project Structure

trtc-ai-tts/
├── server.js           # Main server file with all backend logic
├── conversation.html   # Web interface for conversations
├── package.json        # Dependencies and scripts
├── .env               # Environment variables (not committed)
└── .gitignore         # Git ignore rules

Usage

1. Start a conversation: Click "Start Chat" in the web interface
2. Speak or type: The AI will respond with both text and voice
3. Interrupt: Use the interrupt button to stop the AI mid-response
4. End conversation: Click "End Chat" to terminate the session

Development

Run in development mode with auto-reload:

npm run dev

Dependencies

• express - Web framework
• cors - CORS middleware
• dotenv - Environment variable management
• tencentcloud-sdk-nodejs-trtc - Tencent Cloud TRTC SDK
• tls-sig-api-v2 - TRTC signature generation
• morgan - HTTP request logger

License

MIT

Support

For issues and questions, please create an issue in the GitHub repository.