Cheap ‘smart glove’ translates sign language into text

Scientists at the University of San Diego have developed a prototype glove that is able to convert hand gestures into text using built-in sensors.

The researchers built the smart glove, dubbed The Language of Glove, for less than $100, using inexpensive electronics and stretchy materials. The glove can wirelessly convert the manual alphabet of the American Sign Language into letters on the screen of a smartphone or computer via a bluetooth connection.

Each glove has nine stretchable sensors where the knuckles of a user’s fingers are located: two on each finger and one on the thumb. The sensors are made of thin strips of polymers with a conductive layer of carbon. The sensors are mounted on the glove via copper adhesive tape. Connectors connect each sensor to a printed circuit board on the back of the glove.

The device works by making hand gestures. If the glove and thus the sensors are bent or stretched, the electrical resistance of the sensors changes. Based on this, different letters can be binary programmed based on the position of the sensors. A resting knuckle is programmed as a 0; a knuckle in flexed position is coded as a 1.

The glove creates a binary code of nine ones and zeros for each gesture, which represents a certain letter from the manual alphabet. The smart glove contains an accelerometer and a pressure sensor, so that it can also distinguish between, for example, the letter i and the letter j; these letters get exactly the same nine-digit binary code, so to be able to distinguish, it takes more than just measuring the position of the nine sensors.

The scientists are working on further developing the device so that it can also be used for many other applications, such as virtual reality, augmented reality, medical applications or in games. There are already similar gloves that can translate sign language into speech and text, such as the EnableTalk gloves or the SignLanguageGlove.