MCP.so
Sign In

Supertonic3 Mcp

@nextic-tech

About Supertonic3 Mcp

Local, on-device text-to-speech for Claude & Cursor. No API key, no cloud. Powered by Supertonic 3 — 10

Basic information

Category

Developer Tools

Transports

stdio

Publisher

nextic-tech

Submitted by

Sambit Mahapatra

Config

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

{
  "mcpServers": {
    "supertonic3": {
      "command": "uvx",
      "args": [
        "supertonic3-mcp"
      ]
    }
  }
}

Tools

3

Convert text to speech and save a WAV file. Returns the absolute path and metadata.

List built-in Supertonic voice styles.

List inline expression tags (e.g. <laugh>) usable inside speak() text.

Overview

What is Supertonic3 Mcp?

Supertonic3 Mcp wraps Supertonic 3 on-device text-to-speech as three MCP tools your AI assistant can call directly. It requires no API key and runs entirely locally, with model weights cached on first use.

How to use Supertonic3 Mcp?

Clone the repository, install with pip install -e ., optionally run supertonic3-mcp preload to pre-download the model, then add the server to your MCP client configuration using the absolute path to the supertonic3-mcp binary.

Key features of Supertonic3 Mcp

  • No API key or cloud calls required
  • 10 built-in voices: M1–M5 and F1–F5
  • Supports 31 language codes including en, ko, ja, fr
  • Inline expression tags: <laugh>, <pause>, <breath>, and more
  • ~820ms warm synthesis speed on Apple M3 Air
  • Uses FastMCP with asyncio for serialized inference

Use cases of Supertonic3 Mcp

  • Give an AI assistant a voice for spoken responses
  • Synthesize text to WAV files with file path and metadata
  • Integrate on-device TTS into MCP-compatible development tools

FAQ from Supertonic3 Mcp

Comments

More Developer Tools MCP servers