Gesture based Sign Language Recognition System

The use of hand gestures in sign language has been a crucial method of non-verbal communication, particularly for individuals with hearing or speech impairments. Despite numerous sign language systems developed by various developers worldwide, most of them are not flexible or cost-effective for end-users. This project proposes a hand gesture recognition system that can automatically detect sign language, allowing for more effective communication between deaf and dumb people and normal individuals. Gesture and pattern recognition are rapidly evolving fields that play a crucial role in nonverbal communication. Hand gestures have become an integral part of human’s daily lives, and the Hand Gesture recognition system offers a natural and approachable way of communicating with computers, which people are more familiar with. The proposed system combines several cutting-edge technologies such as MobileNet, Xception, ResNet 101, and DenseNet 121. Among the four algorithms, DenseNet 121 achieved the highest overall accuracy, with an accuracy of 98.9%. Real-time sign language detection with text-to-speech technology and the development of a Telegram bot for sign language detection are additional features of this project. The integration of these technologies has resulted in a novel and effective method of sign language communication that can enhance the lives of individuals with hearing or speech impairments.