Voice Generation MCP Server

A Model Context Protocol (MCP) server that provides voice generation capabilities using the Minimax AI API. This server converts text to speech and automatically uploads the generated audio files to Amazon S3 for easy access and sharing.

Features

• Text-to-Speech Generation: Convert text to high-quality speech using Minimax AI's voice synthesis API
• S3 Integration: Automatically upload generated audio files to Amazon S3 with organized directory structure
• MCP Protocol Support: Full compatibility with Model Context Protocol for seamless integration with AI assistants
• Authentication: Built-in API key authentication for secure access
• Multiple Transport Modes: Support for HTTP, SSE, and STDIO transport protocols
• Docker Support: Easy deployment with Docker and Docker Compose
• Configurable Audio Settings: Customizable sample rate, bitrate, and format options

Prerequisites

• Python 3.8 or higher
• Minimax AI API credentials
• Amazon S3 bucket and credentials
• (Optional) Docker and Docker Compose for containerized deployment

Installation

Local Installation

1. Clone the repository

   git clone <repository-url>
   cd voice-gen-mcp
   

2. Create a virtual environment

   python3 -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate
   

3. Install dependencies

   pip install -r requirements.txt
   

4. Configure environment variables

   cp env.example .env
   # Edit .env with your actual configuration values
   

Docker Installation

1. Build the Docker image

   docker build -t voice-gen-mcp .
   

2. Run with Docker Compose

   cp env.example .env
   # Edit .env with your configuration
   docker-compose up -d
   

Configuration

Environment Variables

Create a .env file based on env.example with the following required variables:

Voice Generation API (Required)

VOICE_GEN_API_GROUP_ID=your_minimax_group_id
VOICE_GEN_API_KEY=your_minimax_api_key

S3 Configuration (Required)

S3_BUCKET_NAME=your_s3_bucket_name
S3_REGION=us-east-1
S3_ACCESS_KEY_ID=your_s3_access_key_id
S3_SECRET_ACCESS_KEY=your_s3_secret_access_key
S3_ENDPOINT=https://s3.amazonaws.com
S3_PREFIX=voice-gen/

Usage

Starting the Server

Local Development

python3 server.py

Docker

docker run -d \
  --name voice-gen-mcp \
  -p 8000:8000 \
  --env-file .env \
  voice-gen-mcp

Docker Compose

docker-compose up -d

MCP Clients

The server supports multiple transport modes:

• HTTP: http://localhost:8000/mcp
• SSE: http://localhost:8000/sse
• STDIO: Direct process communication

Available Tools

generate_voice

Converts text to speech and uploads to S3.

Parameters:

• text (string, required): The text to convert to speech
• model (string, optional): Model to use (default: "speech-2.5-hd-preview")
• voice_id (string, optional): Voice ID to use (default: "mylxsw_voice_1")
• speed (float, optional): Speech speed (default: 1.0, typically 0.5-2.0)

Returns:

• Success message with S3 URL and file size
• Error message if generation fails

Example:

{
  "text": "Hello, this is a test of the voice generation system.",
  "model": "speech-2.5-hd-preview",
  "voice_id": "mylxsw_voice_1",
  "speed": 1.2
}

Speed Control:

• speed = 0.5: Half speed (slower speech)
• speed = 1.0: Normal speed (default)
• speed = 1.5: 1.5x speed (faster speech)
• speed = 2.0: Double speed (very fast speech)

License

MIT License