# Quick-TTS MCP Server

MCP server providing text-to-speech functionality using OpenAI's TTS API via the [quick-tts](https://github.com/diogoseca/quick-tts) package.

## Installation

### Using uvx (Recommended)
```bash
uvx quick-tts-mcp
```

### Using pip
```bash
pip install quick-tts-mcp
```

### Using Docker
```bash
docker run -e OPENAI_API_KEY=your-key quick-tts-mcp
```

## Configuration

Create a `.env` file:
```bash
OPENAI_API_KEY=sk-your-openai-api-key-here
```

## Usage

### Start the server
```bash
quick-tts-mcp
```

### Available tools:
- `generate_speech`: Convert text to speech
- `list_voices`: List available voices  
- `list_models`: List available models

## Example MCP client configuration
```json
{
  "mcpServers": {
    "quick-tts": {
      "command": "uvx",
      "args": ["quick-tts-mcp"],
      "env": {
        "OPENAI_API_KEY": "sk-your-key"
      }
    }
  }
}
```

## Tools

### generate_speech
Convert text to speech using OpenAI's TTS API.

**Parameters:**
- `text` (required): Text to convert to speech
- `voice` (optional): Voice to use - alloy, echo, fable, onyx, nova, shimmer (default: alloy)
- `model` (optional): Model to use - tts-1, tts-1-hd (default: tts-1-hd)
- `output_format` (optional): Output format - mp3, wav (default: mp3)

**Returns:**
JSON with file path, size, and metadata

### list_voices
List all available TTS voices with descriptions.

**Returns:**
JSON array of available voices

### list_models
List all available TTS models with descriptions.

**Returns:**
JSON array of available models

## Development

### Local installation
```bash
pip install -e .
```

### Running locally
```bash
python -m quick_tts_mcp.server
```

## License

MIT