Signify is an innovative capstone project introducing a real-time accessibility system that translates speech and text into sign language via a dynamic avatar. The project aims to bridge the communication gap for the deaf and hard-of-hearing community by making spoken content instantly accessible.

A core engineering focus is maintaining ultra-low latency across the entire processing pipeline. The system utilizes a double audio capture mechanism to simultaneously record internal system audio and microphone input. This audio is rapidly transcribed using the Google STT API and processed through an advanced AWS cloud architecture. A distributed workflow then extracts precise kinematics using Mediapipe models, which OpenCV seamlessly renders into fluid animations.

Crucially, Signify integrates natively with platforms like Zoom, Teams, and Google Meet, delivering a highly scalable, real-time accessibility solution tailored for modern digital workspaces.