A real-time audio chat application using OpenAI's realtime audio API with WebRTC. Ask about the weather in any location and get real-time responses using Open-Meteo API.

Notes:

• work in progress
• works with OpenAI and Azure OpenAI; by default Azure OpenAI is used
• for Azure OpenAI, deploy the service in Sweden Central and deploy the gpt-4o-realtime-preview model
• WebRTC API is in preview

Check .env.example for an example .env file.

• Real-time audio streaming
• Live transcription
• Weather function integration with Open-Meteo API
• Google search integration (requires SERPER_API_KEY)
• WebRTC communication
• FastAPI backend to get a ephemeral session token and to get weather data

1. Clone the repository
2. Create a virtual environment: python -m venv .venv
3. Activate it:
   • Windows: .venv\Scripts\activate
   • Unix/macOS: source .venv/bin/activate
4. Install dependencies: pip install -r requirements.txt
5. Create .env file based on .env.example

Notes:

• Go to https://serper.dev/ and get your API key.
• Go to https://openai.com and get your API key.
• Modify app.py to use the OpenAI API instead of the Azure OpenAI key
• Modify config.js to use the OpenAI realtime endpoint instead of the Azure OpenAI realtime endpoint
• Modify the model name in the /session endpoint in app.py to use the OpenAI model name instead of the Azure OpenAI model name

1. Start server: python app.py
2. Open index.html in a browser (Tip: use live server extension for VSCode)
3. Click Start and allow microphone access
4. Try asking: "What's the weather like in Amsterdam?"

• app.py: FastAPI backend server
• index.html: Frontend interface
• config.js: Configuration for the frontend
• app.js: Frontend logic
• requirements.txt: Python dependencies
• test.http: API endpoint tests
• .env: Environment variables (create this)