Text-to-Speech MCP Server

A simple MCP (Model Context Protocol) server that provides text-to-speech conversion capabilities, written in TypeScript.

Project Purpose

This project implements a Model Context Protocol (MCP) server that offers text-to-speech functionality. It allows AI assistants and other MCP clients to convert text into speech audio files (MP3 format) through a standardized interface.

Key features:

• Convert any text to speech audio
• Save the resulting audio as MP3 files
• Simple integration with any MCP-compatible client

quickstart

install

npm install -g mcp-koko-tts

mcp path

Please run it in the root directory of the project.

current_dir=$(pwd); mcp_path="$current_dir/build/index.js"; echo "MCP Path is set to: $mcp_path"

environment variable

export TTS_OUTPUT_DIR="~/.mcp_tts"

start the tts docker container

gh repo clone remsky/Kokoro-FastAPI

cd Kokoro-FastAPI
# GPU: Requires NVIDIA GPU with CUDA 12.8 support (~35x-100x realtime speed)
cd docker/gpu
docker compose up --build

# CPU: PyTorch CPU inference
cd docker/cpu
docker compose up --build

Configure the Client

Cherry Studio