SuperClonica text to speech

io.github.superclonica-ttsPublished by GitHub account superclonica-ttsv0.1.1Updated Nov 1, 2026

Local text to speech in about 190 voices and 31 languages, offline with no limits.

VerifiedSTDIODesktop onlyMedia & DesignAI & ML

Overview

AI-generated overview

Local text-to-speech server that lets an assistant voice text into audio files on your own computer, offline and without limits, in 31 languages.

What you can do
  • Find voices: search stock and installed voices plus the online catalog, with status and page for each
  • Generate speech: turn text into an mp3, wav or ogg file on this computer, up to 20000 characters per call
  • Manage voices: download catalog voices for offline use and get the purchase link for voices that must be bought
  • Playback and status: play an audio file with the system player and check whether the model is loaded, on which device, and which folders are used
When to use it
Useful when you want an assistant to produce spoken audio locally, such as narration, voice-overs or reading text aloud, without per-use limits or sending text to a cloud service. It suits desktop setups where offline operation and privacy matter.
Requirements
Runs locally as a stdio process on Windows, macOS or Linux with Python 3.10 or newer, started via uvx, pipx or pip. The first start installs the engine and the first speak downloads a voice model of about 400 MB. SUPERCLONICA_API_KEY is optional and only needed to install bought or own voices; SUPERCLONICA_DEVICE selects cpu or gpu. Desktop only; Chinese is not supported.
Before you install
The optional SUPERCLONICA_API_KEY is a secret for installing bought or own voices. buy_voice only returns a purchase link and never pays, but buying a voice happens in your browser and costs money. speak writes audio files and never overwrites an existing file; install_voice downloads voices to disk. Voice catalog lookups go online to superclonica.com.

Installation

In SourceWeft

  1. Open SuperClonica text to speech in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

SuperClonica MCP server

Text to speech for Claude and other MCP clients, running on your own computer: offline after the first run, no limits, 31 languages. It is the local sibling of the cloud server at https://superclonica.com/mcp, with the same tool names. Voices come from the SuperClonica catalog.

Works on Windows, macOS and Linux with Python 3.10 or newer. Chinese is not supported.

Set it up

You need uv, which gives the uvx command. The first start installs the engine, which takes a minute or two. The first speak downloads the voice model (about 400 MB) once.

Claude Desktop. Add this to claude_desktop_config.json and restart the app:

json
{  "mcpServers": {    "superclonica": {      "command": "uvx",      "args": ["superclonica-mcp"],      "env": { "SUPERCLONICA_API_KEY": "sc_..." }    }  }}

If the app cannot find uvx, put the full path to it in command. On Windows where uvx prints it, on macOS and Linux which uvx.

Claude Code:

bash
claude mcp add superclonica -e SUPERCLONICA_API_KEY=sc_... -- uvx superclonica-mcp

Cursor (~/.cursor/mcp.json) takes the same JSON as Claude Desktop. VS Code (.vscode/mcp.json) takes the same entry under "servers", with "type": "stdio" added.

Without uvx, these start the same server: pipx run superclonica-mcp, or python -m superclonica_mcp after pip install superclonica-mcp.

The API key is optional. Stock and free catalog voices need none. Installing a bought or your own voice does; creating a key is free.

Tools

ToolWhat it does
list_voices(query?, limit?)Voices on this computer and, when online, the catalog. Shows for each whether it is installed, yours, free or must be bought, with its page.
speak(text, voice, language, format?, output_path?, include_audio?)Voices the text on this computer and saves mp3, wav or ogg. Gives the file's path and length, and the audio itself if asked. Up to 20000 characters per call. An existing file is never overwritten.
install_voice(voice)Downloads a voice you may use to this computer. A version you already have is not downloaded again.
buy_voice(voice)Gives the link to the voice's page. It never pays: you buy in your browser, then tell the assistant to install it.
play(path)Plays an mp3, wav or ogg file with the system player.
get_usage()Whether the model is loaded and on what device, and which folders are used. The local engine has no limits.

voice is an id from list_voices, a stock name (F1–F5, M1–M5) or an exact name. When a name fits several voices you get the candidates to choose from, never a guess.

To use your own voice file, put its JSON in the voices folder. It goes by its file name.

Buying a voice

  1. Ask for a voice, for example "find a calm narrator". The assistant calls list_voices.
  2. If it must be bought, the assistant calls buy_voice and gives you a link that opens the purchase window.
  3. Confirm the purchase in your browser. Only your click can buy.
  4. Say "install it". After that the voice works offline.

Settings

All are optional environment variables.

VariableMeaningDefault
SUPERCLONICA_API_KEYKey for installing bought and your own voicesnone
SUPERCLONICA_VOICES_DIRFolder of voicesWindows %APPDATA%\SuperClonica\voices; macOS ~/Library/Application Support/SuperClonica/voices; Linux ~/.local/share/superclonica/voices
SUPERCLONICA_OUTPUT_DIRFolder for audioaudio beside the voices folder
SUPERCLONICA_DEVICEcpu or gpuGPU if the GPU engine is installed, else CPU

Faster on an NVIDIA GPU

uvx superclonica-mcp installs the CPU engine, and the CPU and GPU engines cannot share one environment. For the GPU engine (Windows or Linux x86-64, Python 3.11+), make an environment of its own:

bash
uv venv superclonica-gpu --python 3.12uv pip install --python superclonica-gpu superclonica-gpu-cudauv pip install --python superclonica-gpu --no-deps superclonica-mcpuv pip install --python superclonica-gpu "mcp>=2.3,<3" "truststore>=0.10"

Then use the Python of that environment as the command, with -m superclonica_mcp as the arguments and SUPERCLONICA_DEVICE=gpu in env.

When something goes wrong

  • "The voice model is still loading": the first run is downloading the model, and that goes on in the background. Ask again in a minute.
  • The client shows no tools: look at the server's log (Claude Desktop: Settings → Developer). The server writes its log only to stderr.
  • A long text times out: on a CPU, voice long texts in parts.

License

Apache-2.0. The voice model is Supertonic 3 under its own OpenRAIL-M license, downloaded from its original place by the engine.

Source: README.md at commit 163851f

Tools

6

list_voices

Find a voice

Search the voices on this computer (stock and installed) and, when online, the catalog at superclonica.com. Each voice shows whether it is installed, bought, free or must be bought, with its page. Each voice speaks all 31 languages; it sounds most natural in the language it was recorded in.

Parameterslimitquery

speak

Voice a text

Turn a text into speech on this computer and save an audio file: offline, without limits. Only installed and stock voices speak. Up to 20000 characters per call; a long text takes longer, so voice long texts in parts.

Parameterstextrequiredvoicerequiredformatlanguagerequiredoutput_pathinclude_audio

install_voice

Install a voice

Download a voice from superclonica.com to this computer, after which speak can use it offline. Free voices need no key; bought and own voices need SUPERCLONICA_API_KEY. A voice not bought yet is not installed: use buy_voice.

Parametersvoicerequired

buy_voice

Page to buy a voice

Give the link to the page where the person buys a catalog voice in their browser. This tool never pays and cannot spend money. When the person says it is bought, call install_voice.

Parametersvoicerequired

play

Play an audio file

Play an mp3, wav or ogg file on this computer with the system player. Returns as soon as playback starts.

Parameterspathrequired

get_usage

Engine status and limits

The local engine has no limits: this tells whether the voice model is loaded, on what device, and which folders are used.

Takes no parameters.

Version history

1
  1. v0.1.1LatestNov 1, 2026

Score

90% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts
  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listedRequired15 pts
  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts