American Sign Language Translation through Sensory Glove; SignSpeak

Abstract

To make a communication bridge, a highly accurate, cost effective and an independent glove was designed for deaf/mute people to enable them to communicate. The glove translates the sign language gestures into speech according to the American Sign Language Standard. The glove contained flex and contact sensors to detect the movements of the fingers and bending of the palm. In addition an accelerometer was built in the glove to measure the acceleration produced by the changing positions of the hand. Principal Component Analysis (PCA) was used to train the glove into recognizing various gestures, and later classify the gestures into alphabets in real time. The glove then established a Bluetooth link with an Android phone, which was used to display the received letters and words and convert the text into speech. The glove was found to have an accuracy of 92%.