Simple Voice

Provides text-to-speech capabilities using multiple character voices with simultaneous playback and a custom pronunciation dictionary.

python

0

GitHub Stars

python

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "kurasuai-inc-voice-mcp": {
      "command": "uv",
      "args": [
        "--directory",
        "/path/to/voice-mcp",
        "run",
        "simple_voice_mcp.py"
      ],
      "env": {
        "VOICE_MODEL": "syouzyo_4",
        "VOICE_API_BASE": "https://your-custom-api.com"
      }
    }
  }
}

You can run a simple voice MCP server that converts text into spoken audio, supporting multiple character voices and simultaneous playback. This makes it easy to add voice interaction to your projects without writing complex audio pipelines.

How to use

Set up an MCP client to connect to the simple voice MCP server family. You will run a local MCP server process via the UV tool, point it at your voice MCP directory, and then send text to be spoken. You can choose different character voices and run multiple servers at once to handle different speaker personalities.

Example usage patterns you can implement after setup include: 1) Speaking with a specific character by selecting a model, 2) Playing multiple voices simultaneously, and 3) Extending pronunciation for English words using a custom dictionary.

{
  "mcpServers": {
    "simple_voice": {
      "command": "uv",
      "args": [
        "--directory", "/path/to/voice-mcp", 
        "run", 
        "simple_voice_mcp.py"
      ]
    },
    "simple_voice_sutera": {
      "command": "uv",
      "args": [
        "--directory", "/path/to/voice-mcp", 
        "run", 
        "simple_voice_mcp.py", 
        "--model", 
        "sutera"
      ]
    }
  }
}

Notes on models and dictionary

You can switch the voice model by adding a --model option to the server configuration. The available voice models include a wide range of male, female, and child voices. You can also register custom word pronunciations in a shared dictionary to control how English words are read.

Custom dictionary usage

Use the built-in dictionary tools to add, remove, and list pronunciations for English words. These changes will be reflected in real time across MCP processes that share the dictionary.

Available tools

say

Text-to-speech entry point that plays spoken output using the configured character voice

add_to_dictionary

Register new English words with their Katakana readings in the custom dictionary for pronunciation control

remove_from_dictionary

Remove words from the custom dictionary to revert to default pronunciations

list_dictionary

Display all registered words and their readings in the custom dictionary