Rime Text To Speech

🚀 Rime MCP English Version

A Model Context Protocol (MCP) server that uses the Rime API to provide text-to-speech functionality and plays the audio through the system's native audio player.

🚀 Quick Start

Rime MCP is a server that utilizes the Rime API to convert text into speech and plays the generated audio via the system's native audio player.

✨ Features

• Provide a speak tool to convert text into speech and output it through the system audio.
• Use Rime's high-quality speech synthesis API.

📦 Installation

Configuration Requirements

• Node.js version 16 or higher.
• A properly functioning audio output device.
• macOS: Use afplay.

Although not tested, sample code provided by Claude is available:

• Windows: Built-in Media.SoundPlayer (PowerShell).
• Linux: mpg123, mplayer, aplay, or ffplay.

MCP Configuration

"ref": {
  "command": "npx",
  "args": ["rime-mcp"],
  "env": {
      RIME_API_KEY=your_api_key_here

      # Optional configuration
      RIME_GUIDANCE="<Guide the agent on how to use the speak tool>"
      RIME_WHO_TO_ADDRESS="<Name to address>"
      RIME_WHEN_TO_SPEAK="<Tell the agent when to speak>"
      RIME_VOICE="cove" 
  }
}

All optional environment variables are part of the tool's definition and are used for prompting.

All voice options are listed here.

You can obtain your API key on the Rime dashboard.

The following environment variables can be used to customize behavior:

• RIME_GUIDANCE: The main description of using the speak tool.
• RIME_WHO_TO_ADDRESS: The object to address in the voice (Default: "User").
• RIME_WHEN_TO_SPEAK: When to use the tool (Default: "When the command is completed or when asked to speak").
• RIME_VOICE: The default voice to use (Default: "cove").

💻 Usage Examples

Video Demo

Example 1: Agent Announces Code Changes

"RIME_WHEN_TO_SPEAK": "Always make a voice announcement at the end of the answer.",
"RIME_GUIDANCE": "Briefly summarize the content of the answer. If files are edited, list them."

Example 2: Learn How Young People Speak

RIME_GUIDANCE="Use phrases and slang commonly used by millennials."
RIME_WHO_TO_ADDRESS="Matt"
RIME_WHEN_TO_SPEAK="When asked to speak"

Example 3: Select Different Languages Based on Context

RIME_VOICE="Use 'cove' when discussing Typescript and 'antoine' when talking about Python"

🚧 Development Steps

1. Install dependencies:

npm install

2. Build the server:

npm run build

3. Run the development mode with hot reloading:

npm run dev

📄 License

This project is licensed under the MIT license.