ADA v1.0 (First Release)

V1.0 is ADA's first version. ADA is a modular personal assistant that combines facial recognition, wake word detection, and GPT-4o-powered visual analysis to create personalized user experiences. The system activates via phrases like "Hey ADA" and can answer questions about what it sees through the camera. Built with Python 3.10+, ADA features user recognition for customized interactions, natural speech processing, and real-time visual feedback. The architecture includes six main modules (User Detection, Activator, TTS, User Input, Vision GPT, and Models) working together to provide a seamless experience. Setup requires an OpenAI API key and basic hardware with a webcam and microphone.

See the model architecture below, more images and demos too come.