MCP Tts Server

🚀 mcp-tts-server

This project offers a Text-to-Speech (TTS) server based on the Model Context Protocol (MCP), utilizing the Groq platform for speech generation.

🚀 Quick Start

The mcp-tts-server provides a TTS solution based on the MCP protocol. It connects to the Groq API to generate realistic speech audio and offers various useful features and tools.

✨ Features

• Implements the TTS function of the MCP protocol.
• Connects to the Groq API (https://api.groq.com/openai/v1/audio/speech) to generate realistic speech audio.
• Provides the generate_speech tool, accessible via the MCP protocol.
• Supports multiple audio output formats (wav, mp3, aac, opus, flac).
• Uses the @tropicbliss/symphonia library to play the generated audio directly on the server side (optional feature).
• Requires setting the Groq API key (GROQ_API_KEY) through environment variables.

📦 Installation

Run the following command to install all necessary dependencies:

bun install

💻 Usage Examples

Basic Usage

To run the server using the standard input-output transmission protocol, add the following content to the configuration file:

{
  "mcpServers": {
    "mcp-tts-server": {
      "command": "bunx",
      "args": [
        "--bun",
        "mcp-tts-server"
      ],
      "env": {
        "GROQ_API_KEY": "gsk_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx"
      }
    }
  }
}

📚 Documentation

Project Structure

• src/server.ts: Defines the core logic of the MCP server, including the generate_speech tool and the interaction with the Groq API.
• src/cli.ts: Provides a standard console interface for running the MCP server.