This guide walks you through setting up the Voice AI Agent locally.
1️⃣ Install uv (Python Package Manager)
This project uses UV Package Manager for fast, reproducible Python environments.
pip install uvVerify installation:
uv --version2️⃣ Clone the Repository
git clone https://github.com/CyberXpert607/Voice-AI-Agents.gitcd Voice-AI-Agents3️⃣ Sync the Python Environment
This project does NOT require requirements.txt.
Instead, dependencies are managed via:
pyproject.toml
uv.lock
Install dependencies
uv syncThis will:
Create a virtual environment automatically
Install exact, locked dependency versions
4️⃣ Create a .env File
Create a .env file in the project root:
cp .env.example .envOr manually create one if .env.example is not present. Also make sure to use the command:
cd src/
uv run python agent.py download-filesto install the dependencies needed for the agent to run.
5️⃣ Obtain Required API Keys
This agent requires multiple external services.
🔹 OpenAI (LLM)
Used for reasoning and conversation.
Go to: https://platform.openai.com/
Create an account
Generate an API key
Copy the key
🔹 Groq (LLM / optional TTS)
Used for ultra-fast inference.
Go to: https://console.groq.com/
Create an account
Generate an API key
Copy the key
🔹 ElevenLabs (Text-to-Speech)
Used for high-quality voice synthesis.
Go to: https://elevenlabs.io/
Create an account
Generate an API key
Copy the key
🔹 LiveKit (Realtime Audio / Rooms)
Go to: https://cloud.livekit.io/
Create a project
Copy:
API Key
API Secret
WebSocket URL
6️⃣ Configure Environment Variables
Open your .env file and fill in the values:
Obviously here you can choose which model to use for your agent either openai, or google, or anthropic, it's really your choice. OPENAI_API_KEY=your_openai_api_key_here
GROQ_API_KEY=your_groq_api_key_here
ELEVENLABS_API_KEY=your_elevenlabs_api_key_here
LIVEKIT_API_KEY=your_livekit_api_key_here
LIVEKIT_API_SECRET=your_livekit_api_secret_here
LIVEKIT_URL=wss://your-livekit-url.livekit.cloud
OPENWEATHER_API_KEY=your_open_weather_api_key
7️⃣ Run the Agent
Activate the environment and start the agent:
cd src/
uv run python agent.py console