TextToSpeech-MCP Server

@SmartLittleApps

About TextToSpeech-MCP Server

Local text-to-speech MCP server with dual engines (macOS Say + Kokoro TTS) for Claude Desktop

#text-to-speech #local-tts #speech-synthesis

Config

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

{
  "mcpServers": {
    "local-tts-mcp": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "node",
        "dist/server.js"
      ]
    }
  }
}

Tools

No tools detected

We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.

Overview

What is TextToSpeech-MCP Server?

TextToSpeech-MCP Server is a local MCP (Model Context Protocol) server that provides privacy-first text-to-speech synthesis using macOS Say and Kokoro TTS engines. Designed for MCP-compatible clients like Claude Desktop, it keeps all processing on your machine.

How to use TextToSpeech-MCP Server?

Install dependencies with npm install, build with npm run build, and test with npm start. For Claude Desktop, add the provided JSON configuration to your settings. Use tools like synthesize_text, list_voices, health_check, and play_audio to convert text to speech, manage voices, and control playback.

Key features of TextToSpeech-MCP Server

All TTS synthesis happens locally – no data leaves your machine
Two engines: macOS Say (23 voices) and Kokoro TTS (21 voices)
Supports 43 languages including English, Spanish, French, Japanese, Chinese
Customizable output formats: aiff, wav, mp3, m4a
Adjustable speech speed (0.1–3.0) and quality (fast, balanced, high)

Use cases of TextToSpeech-MCP Server

Convert simple text to speech using the default voice
Select a specific voice by language, gender, or engine for natural output
Synthesize multilingual content, such as "Bonjour le monde" with a French voice

FAQ from TextToSpeech-MCP Server

What TTS engines does TextToSpeech-MCP Server support?

It supports macOS Say (23 voices, multiple languages) and Kokoro TTS (21 voices, 8 languages), with automatic engine selection.

What are the system requirements?

macOS (for the macOS Say engine), Node.js 18+, and TypeScript.

Is my data sent to external servers?

No. All text-to-speech synthesis happens locally; no data is sent to external services.

What is the maximum text length for synthesis?

The text parameter supports up to 50,000 characters.

How do I configure TextToSpeech-MCP Server for Claude Desktop?

Add the provided JSON configuration to your Claude Desktop settings, specifying the command node ./dist/server.js and the correct working directory.

Frequently asked questions