mcpField
S
Supertonic3 Mcp

Supertonic3 Mcp

Local
github.com/nextic-techDeveloper Tools0Updated 19w ago
Repository

About Supertonic3 Mcp

supertonic3-mcp is a local text-to-speech MCP server for Claude, Cursor, and other MCP clients. It uses Supertonic 3 on-device, with no API key or cloud service, and can preload model files for offline synthesis.

Key features

  1. 01Synthesize text to WAV
  2. 02List built-in voices
  3. 03List expression tags
  4. 04Run locally over STDIO
  5. 05Preload models for offline use

Use cases

  1. 01Generate speech from text in Claude
  2. 02Create WAV output from Cursor
  3. 03Synthesize speech without network access

MCP support

The STDIO MCP server exposes `speak`, `list_voices`, and `list_expressions` to MCP clients. `speak` returns the WAV file path and metadata.

#tts#text-to-speech#audio#voice#local-ai#on-device#no-api-key#apple-silicon#supertonic#fastmcp

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "supertonic3": {
      "command": "uvx",
      "args": [
        "supertonic3-mcp"
      ]
    }
  }
}
Run commanduvx supertonic3-mcp

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
pypisupertonic3-mcp

Tools

3

Saved tool list · updated 2w ago

ParameterTypeRequiredDescription
languageanyNoISO 639-1 code (en, ko, ja, …). Required for reliable non-English output.
playbooleanNoIf true, play audio on this machine (STDIO/local use only).
speednumberNoPlayback speed in [0.7, 2.0].
textstringYes1–5000 characters; may include expression tags from list_expressions().
voice_idanyNoBuilt-in voice (M1–M5, F1–F5). See list_voices().

Details

From the project README

supertonic3-mcp

Local, on-device TTS for Claude & Cursor, powered by Supertonic 3. No API key. No cloud. An internal tool open-sourced by Halozen — we build AI compliance intelligence for construction.

Not affiliated with Supertone Inc.

Expose speak, list_voices, and list_expressions to Claude Desktop, Cursor, or any MCP client over STDIO.

Quick start (TTHW < 3 min)

git clone https://github.com/nextic-tech/supertonic3-mcp && cd supertonic3-mcp
python3 -m venv .venv && source .venv/bin/activate
pip install -e ".[dev]"

# Optional: pre-download model for offline use (~400MB)
supertonic3-mcp preload

# Run MCP server (STDIO)
supertonic3-mcp
Cursor MCP config

Add to .cursor/mcp.json (or Cursor Settings → MCP):

{
  "mcpServers": {
    "supertonic3": {
      "command": "/absolute/path/to/supertonic-tts/.venv/bin/supertonic3-mcp",
      "args": []
    }
  }
}

First server start downloads the Supertonic model into ~/.cache/supertonic3/ unless you ran preload first.

Tools

ToolDescription
speakSynthesize text to a WAV file; returns absolute path + metadata
list_voicesBuilt-in voices (voice_id, gender)
list_expressionsInline tags (<laugh>, <breath>, …) with descriptions
speak parameters
  • text — 1–5000 characters; expression tags allowed
  • voice_id — optional (M1, F1, …)
  • language — ISO 639-1 (en, ko, ja, …). For non-English text, always set language=. Defaults to en.
  • speed — 0.7 to 2.0 (SDK range)
  • play — if true, plays audio on this machine via afplay (macOS) or aplay (Linux). Unsupported on Windows.

WAV files are written to /tmp/supertonic_*.wav (macOS/Linux). Windows is not supported for synthesis output paths in v1.0.

Example return:

Audio saved to /tmp/supertonic_abc123.wav (1.4s, voice: M1, lang: en)

Performance (this repo)

Measured on Apple M3, supertonic 1.3.1 — see benchmark/results.md.

ScenarioMedian FSL
Warm (model loaded)~0.82s
Cold (new TTS() per call)~0.81s

FSL = time from synthesize() through WAV written (no streaming, no play=True).

Re-run: python benchmark/run.py

Offline use

supertonic3-mcp preload

Downloads ONNX weights atomically to ~/.cache/supertonic3/ and prints SHA256 checksums. After preload, synthesis works without network access.

Development

pip install -e ".[dev]"
pytest

Tests mock the Supertonic SDK (no network in CI).

Coming in v1.1

  • listen() — Whisper speech-to-text (pip install supertonic3-mcp[stt])
  • SSE transport + Docker image for remote agents
  • PyPI publish workflow

License

MIT (this package). Supertonic SDK is MIT; model weights use OpenRAIL-M.

Disclaimer

AI-generated speech is not a substitute for certified safety, legal, or medical guidance. For demonstration purposes only.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.