MCP Elevenlab Scribe ASR

🚀 ElevenLabsscribe MCP Server

An implementation of the Model Control Protocol (MCP) server for ElevenLabs' Scribe speech-to-text API, offering real-time transcription capabilities with advanced context management and bi-directional streaming technology support.

🚀 Quick Start

The ElevenLabsscribe MCP Server is an implementation for ElevenLabs' Scribe speech - to - text API. It provides real - time transcription features and supports advanced context management and bi - directional streaming technology.

✨ Features

• Real - time Transcription: Stream audio directly from the microphone and get instant transcription results.
• Batch File Processing: Upload audio files for batch processing.
• MCP Protocol Support: Fully implement the Model Control Protocol to enhance context management capabilities.
• WebSocket Support: Enable real - time two - way communication.
• Context Management: Maintain conversation context to improve transcription accuracy.
• Multiple Audio Formats: Support multiple audio formats and perform automatic conversion.
• Language Detection: Automatically identify languages and provide confidence scores.
• Event Detection: Identify speech and non - speech audio segments.

📦 Installation

1. Clone the repository:

git clone https://github.com/aromanstatue/MCP-Elevenlab-Scribe-ASR.git
cd MCP-Elevenlab-Scribe-ASR

2. Create and activate a virtual environment:

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

3. Install dependencies:

pip install -e .

4. Create a .env file containing the API key:

ELEVENLABS_API_KEY=your-api-key-here

💻 Usage Examples

Basic Usage

Start the Server

python -m elevenlabs_scribe_mcp_server.main

The server will start on the default port 8000 or the next available port.

Example Client Usage

1. File transcription:

python examples/client_example.py --file path/to/audio.wav

2. Microphone transcription:

python examples/client_example.py --mic

Advanced Usage

API Interfaces

1. REST API:

• POST /transcribe: Upload an audio file for transcription.
• GET /health: Health check interface.

2. WebSocket API:

• ws://localhost:8000/ws/transcribe: Real - time audio transcription.

MCP Protocol

The server implements the Model Control Protocol (MCP) and supports the following message types:

1. INIT: Initialize a new transcription session.
2. START: Start audio streaming.
3. AUDIO: Send audio data.
4. TRANSCRIPTION: Receive transcription results.
5. ERROR: Error message.
6. STOP: End audio streaming.
7. DONE: Complete the session.

📚 Documentation

Development Guide

Run Tests

pytest tests/

Project Structure

elevenlabs-scribe-mcp-server/
├── elevenlabs_scribe_mcp_server/
│   ├── __init__.py
│   ├── main.py              # FastAPI server entry point
│   └── mcp/
│       ├── __init__.py
│       ├── protocol.py      # MCP protocol processing module
│       ├── types.py         # Protocol type definitions
│       └── elevenlabs.py    # ElevenLabs implementation module
├── examples/
│   └── client_example.py    # Example client code
├── tests/
│   └── test_transcribe.py   # Test suite
├── pyproject.toml           # Project metadata file
└── README.md                # Project documentation

Dependencies

• Python 3.8+
• FastAPI
• Uvicorn
• PyAudio (for microphone support)
• aiohttp
• python - dotenv
• pydantic

Contribution Guide

1. Fork the repository.
2. Create a feature branch (git checkout -b feature/amazing-feature).
3. Commit your changes (git commit -m 'Add amazing feature').
4. Push to the branch (git push origin feature/amazing-feature).
5. Submit a Pull Request.

📄 License

MIT License

For more detailed information, please refer to the official documentation.