mcp-registry/kokoro-tts-mcp

    ==================
      
       /// MCP ///
      /// KOK ///
        
    ==================
        
    [server:online]
    [protocol:ready]

kokoro-tts-mcp

by mberg

Kokoro Text-to-Speech MCP server that converts text to MP3 locally and/or uploads the result to S3.

51
12
Open Source
GitHub

01

mcp-tts.py

Server script that hosts the Kokoro Text-to-Speech MCP service, converts supplied text to MP3 audio and can optionally upload the resulting file to S3.

02

mcp_client.py

Command-line client for sending Text-to-Speech requests to the MCP server, supporting direct text input, file input, voice and speed customization, and optional suppression of S3 uploads.

Installation

1. Prerequisites
• Python 3.9 or later
• git and a C/C++ build tool-chain (if native libraries are compiled)
2. Clone the repository
git clone https://github.com/mberg/kokoro-tts-mcp.git cd kokoro-tts-mcp
3. Create and activate a virtual environment (recommended)
python -m venv .venv source .venv/bin/activate # Windows: .venv\Scripts\activate
4. Install runtime dependencies
pip install -r requirements.txt # if the file exists
If no requirements.txt is provided, install the package manually:
pip install .
5. Configuration
• Copy the sample configuration file (e.g., `config.sample.yml`) to `config.yml` and edit values such as the TTS provider, voice, language, and any API keys.
• If the server integrates with Google Cloud TTS, export the path to your GCP service account key:
export GOOGLE_APPLICATION_CREDENTIALS=/path/to/key.json
6. Run the MCP Server
python -m kokoro_tts_mcp.server --host 0.0.0.0 --port 8080
The server should now be reachable at `http://localhost:8080`.
7. (Optional) Docker
docker build -t kokoro-tts-mcp . docker run -p 8080:8080 kokoro-tts-mcp
8. Updating
git pull origin main pip install -U -r requirements.txt

Documentation

License: Apache License 2.0
Updated 7/30/2025