Voice Assistant

A local voice assistant powered by ElevenLabs Conversational AI. Speak to it, and it speaks back using an AI agent you configure on the ElevenLabs dashboard.

Prerequisites

• Python 3.12+
• An ElevenLabs account
• A configured ElevenLabs Agent (see setup below)

Installation

Create and activate a virtual environment:

bash

python3 -m venv venv
source venv/bin/activate

Then install dependencies:

bash

pip install elevenlabs python-dotenv certifi

To deactivate the virtual environment when done:

bash

deactivate

ElevenLabs Setup

1. Go to ElevenLabs → Conversational AI → Agents
2. Click Start from blank to create a new agent
3. Set your desired System prompt and First message in the agent settings
4. Go to the Security tab and enable the System prompt and First message overrides if you want to control them from code
5. Copy your Agent ID from the agent page
6. Go to your profile → API Keys and create a new key with at minimum:
   • Text to Speech → Access
   • ElevenAgents → Write
   • Voices → Read
   • Models → Access

Environment Variables

Create a .env file in the project root:

AGENT_ID=your_agent_id_here
API_KEY=your_api_key_here

Usage

bash

python3 voice_assistant.py

The assistant will start listening immediately. Press Ctrl+C to stop.

Notes

• On macOS, if you encounter SSL certificate errors, run:

bash

  /Applications/Python\3.12/Install\ Certificates.command