What is motion-aware-voice-chat-bot?

Motion Aware Voice Chat Bot is a webcam-connected AI assistant that reacts when someone walks into view. Using the browser's own camera access, it continuously monitors the video feed for movement by comparing pixel changes between frames. When motion is detected, the bot greets the person by voice, and if they stay silent too long, it will prompt them to interact. You can have a spoken conversation with it, ask it visual questions like "how many fingers am I holding up?", and it will reply out loud. The interaction follows a simple four-state cycle: idle (waiting), detected (someone appeared), engaged (active conversation), and nudged (reminding the user to talk). This state machine runs on the backend, while motion detection and speech happen entirely in the browser without any server-side processing for audio. Voice works through the browser's built-in Web Speech API, meaning no separate speech service is needed. Visual questions are sent to an AI vision model, by default GPT-4o-mini, but any OpenAI-compatible endpoint can be configured. The backend is built with Python 3.12 and FastAPI, handling real-time messaging via WebSocket. The frontend is plain HTML, CSS, and JavaScript with no build tools required. Settings like motion sensitivity, voice speed, pitch, and model selection can be adjusted and are saved in browser storage. The project requires a camera-capable browser (Chrome, Firefox, Safari, or Edge) and a Vision API key to run. It is released under the MIT license.