Voice Assistant

Voice Assistant

A simple voice assistant using LiveKit, Cartesia, OpenAI, Silero, and a Multi-Content Processor (MCP) for WhatsApp integration.

Prerequisites

• Python >=3.13
• uv (Python packager and virtual environment manager)

Setup

1. Clone the repository:

   git clone <your-repo-url>
   cd voice-assistant
   

2. Create and activate a virtual environment (using uv):

   uv venv
   source .venv/bin/activate  # On Linux/macOS
   # .\.venv\Scripts\activate    # On Windows
   

3. Install dependencies:

   uv sync
   

4. Set up Environment Variables:

   Create a file named .env in the root of the project and add the following environment variables with your actual values. You can copy the example below:

   LIVEKIT_URL=
   LIVEKIT_API_KEY=
   LIVEKIT_API_SECRET=
   CARTESIA_API_KEY=
   OPENAI_API_KEY=
   

   • LIVEKIT_URL: Your LiveKit server URL.
   • LIVEKIT_API_KEY: Your LiveKit API key.
   • LIVEKIT_API_SECRET: Your LiveKit API secret.
   • CARTESIA_API_KEY: Your Cartesia API key for Text-to-Speech.
   • OPENAI_API_KEY: Your OpenAI API key for Speech-to-Text and LLM operations.

5. Configure WhatsApp MCP (Multi-Content Processor):

   This project uses an MCP to interact with WhatsApp. You need to have your WhatsApp MCP server running and configured.

   Update the mcp_config.json file in the root of this project with the correct paths for your MCP setup:

   {
     "mcpServers": {
       "whatsapp": {
         "command": "{{PATH_TO_UV}}",
         "args": [
           "--directory",
           "{{PATH_TO_SRC}}/whatsapp-mcp/whatsapp-mcp-server",
           "run",
           "main.py"
         ]
       }
     }
   }
   

   • Replace {{PATH_TO_UV}} with the absolute path to your uv executable. You can find this by running which uv (on Linux/macOS) or where uv (on Windows) in your terminal.
   • Replace {{PATH_TO_SRC}} with the absolute path to the directory containing your whatsapp-mcp project. For example, if your whatsapp-mcp project is in /Users/me/dev/whatsapp-mcp, then this value should be /Users/me/dev. The final path in the config will be /Users/me/dev/whatsapp-mcp/whatsapp-mcp-server.

Running the Assistant

1. (Optional) Download necessary model files (e.g., for Silero VAD): The voice.py script might have a command or automatically download necessary models on first run. The example below shows a hypothetical command if your voice.py supported it.

   # If your voice.py has a download command, e.g.:
   # uv run python voice.py download-files
   

   Currently, the voice.py script uses livekit.agents.cli.run_app, which typically handles its own setup for plugins like Silero.

2. Start the voice assistant:

   uv run voice-assistant-cli
   

   This uses the voice-assistant-cli script defined in pyproject.toml.

Connecting to the Playground

Once the assistant is running, you can load the LiveKit Agents Playground in your browser and connect to your local agent.

---

Note: This README assumes you have uv installed and configured in your system PATH."# Whatsapp-mcp-voice-agent"