Indian Sign Language Interpreter Using AI

We herein introduce a real-time Indian Sign Language (ISL) interpreter system based on deep learning that translates hand gestures to human-legible and hearable representations. Using MediaPipe for landmark detection and a Convolutional Neural Network (CNN) developed using TensorFlow/Keras, the system detects ISL handshapes that correspond to letters (A–Z) and numbers (0–9). The predictions are rendered in a real time using a Flask-based website, and optionally spoken out using Google Text-to-Speech (gTTS). The tool can be a valuable assistant for those who use ISL by making it easy to communicate across digital channels. The system is intended to be lightweight, deployable, and easy to use, thus fitting for educational, healthcare, and accessibility assistive uses.