Electrical and Computer Engineering

ECE22: Single Hand Real-Time ASL Alphabet & Numbers to Text Translator and Trainer

A real-time, browser-based ASL translator that uses hand tracking and neural networks to convert alphabet and number signs into

Project Description:

This project is a real-time, browser-based American Sign Language (ASL) translator that converts hand signs for the alphabet (A–Z) and numbers (0–10) into text using computer vision and neural networks. It leverages MediaPipe for hand landmark detection and a custom 100-dimensional feature pipeline to enable fast and accurate classification directly in the browser with no server or installation required. The system achieves over 99% accuracy while maintaining real-time performance, and also includes learning and practice modes to help users improve their ASL skills, making it both a translation tool and an educational platform. ASL Fingerspelling Translator

Advisor/Instructor:

Dr. Jerry Wu

Team Members:

Nishanth Sasikumar	Electrical and Computer Engineering

Table #:

C22