SignSpeakAI is a browser-based accessibility prototype that captures webcam video, detects hand landmarks using MediaPipe Hands (Web), classifies gestures using a lightweight in-app “Neural Engine”, and outputs text, emoji, or synthesized voice in real time.