MCP.so
Sign In
L

Local Voice Mcp

@CodeCraftersLLC

About Local Voice Mcp

Give your MCP clients the ability to speak by running local voice models

Config

Add this server to your MCP-compatible client using the configuration below.

{
  "mcpServers": {
    "local-voice-mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@codecraftersllc/local-voice-mcp"
      ],
      "env": {
        "USE_MALE_VOICE": "false",
        "CHATTERBOX_EXAGGERATION": "0.5",
        "CHATTERBOX_CFG_WEIGHT": "1.2",
        "CHATTERBOX_MAX_CHARACTERS": "2000",
        "CHATTERBOX_PLAYBACK_VOLUME": "100"
      }
    }
  }
}

Tools

3

Convert text to speech using the configured TTS engine (Chatterbox Turbo or Kokoro). For Chatterbox Turbo, use paralinguistic tags directly in text for expressive speech: [laugh], [sigh], [cough], [chuckle], [gasp], [groan], [clear throat], [sniff], [shush]. Supports voice cloning via reference audio.

Play an audio file using the system's default audio player with optional volume control and file cleanup. Supports WAV and MP3 formats. Files must be in the temporary directory for security.

Get the current status of the TTS service, including operational state and capabilities.

Overview

What is Local Voice MCP?

Local Voice MCP is a Model Context Protocol server that enables AI clients to speak by running local voice models using Chatterbox TTS. It can operate in either MCP server (stdio) or HTTP server mode, making it suitable for developers who want private, on-device text-to-speech within MCP-compatible editors and tools.

How to use Local Voice MCP?

Install globally via npm install -g local-voice-mcp or run directly with npx -y @codecraftersllc/local-voice-mcp. Configure your MCP client with the command local-voice-mcp-server (or npx variant) and optional environment variables. Use the tools synthesize_text, play_audio, and tts_status to generate and play speech, or run in HTTP mode with MCP_MODE=http.

Key features of Local Voice MCP

  • Full MCP server using @modelcontextprotocol/sdk
  • ElevenLabs-compatible REST API
  • High-quality TTS via Chatterbox TTS
  • Voice cloning with reference audio
  • Adjustable prosody controls
  • Dual mode: MCP or HTTP server

Use cases of Local Voice MCP

  • Enable AI coding assistants like Cursor to speak responses aloud
  • Generate

Comments

More Other MCP servers