Kokoro TTS

🚀 Kokoro TTS MCP Server

A Model Context Protocol (MCP) server that provides text-to-speech (TTS) functionality using the Kokoro TTS engine. This server exposes the TTS functionality through the MCP tool, allowing you to easily integrate speech synthesis into your applications.

🚀 Quick Start

This server offers a seamless way to incorporate text-to-speech capabilities into your projects. Before you start, ensure you meet the prerequisites and follow the installation steps.

✨ Features

• Customizable text-to-speech synthesis with various voices.
• Adjustable speech speed for a personalized experience.
• Support for saving audio to a file or playing it directly.
• Cross-platform audio playback support on Windows, macOS, and Linux.

📦 Installation

Prerequisites

• Python 3.10 or higher
• uv package manager

Steps

1. First, install the uv package manager:

curl -LsSf https://astral.sh/uv/install.sh | sh

2. Clone this repository and install the dependencies:

uv venv
source .venv/bin/activate  # On Windows, use: .venv\Scripts\activate
uv pip install .

💻 Usage Examples

Basic Usage

The server provides an MCP tool generate_speech with the following parameters:

• text (required): The text to be converted to speech.
• voice (optional): The voice to use (default: "af_heart").
• speed (optional): The speech speed multiplier (default: 1.0).
• save_path (optional): The directory to save the audio file.
• play_audio (optional): Whether to play the audio immediately (default: False).

from mcp.client import Client

async with Client() as client:
    await client.connect("kokoro-tts")
    
    # Generate and play speech
    result = await client.call_tool(
        "generate_speech",
        {
            "text": "你好，世界！",
            "voice": "af_heart",
            "speed": 1.0,
            "play_audio": True
        }
    )

📚 Documentation

Dependencies

• kokoro >= 0.8.4
• mcp[cli] >= 1.3.0
• soundfile >= 0.13.1

Platform Support

Audio playback is supported on the following platforms:

• Windows (using start)
• macOS (using afplay)
• Linux (using aplay)

MCP Configuration

Add the following configuration to your MCP settings file:

{
  "mcpServers": {
    "kokoro-tts": {
      "command": "/Users/giannisan/pinokio/bin/miniconda/bin/uv",
      "args": [
        "--directory",
        "/Users/giannisan/Documents/Cline/MCP/kokoro-tts-mcp",
        "run",
        "tts-mcp.py"
      ]
    }
  }
}

📄 License

[Add your license information here]