Researchers from the University of California Los Angeles created a glove that translates American sign language into English simultaneously through a telephone application, to help sign language users address others without an interpreter, and also help people learn sign language.
In each glove, the researchers installed five elastic sensors along each finger, capturing the movement of the hand and the body of the fingers representing letters, numbers, and phrases, and sending them as electric signals to a small electric board above the wrist, to send them to the application for translation.
The application translates into spoken English, at a rate of one word per second, and in the tests they added facial sensors, capturing the eyebrows and lips expressions used in sign language for more accuracy, and published their research in the journal Nature Electronics.
John Chen, assistant professor of biological engineering at the University of Henry Samuili School of Engineering and Applied Sciences of the university, and the principal investigator of this study, said that the previous devices were large, heavy and uncomfortable, and this glove is inexpensive, as it is composed of rubber polymers and flexible sensors, according to an observatory report The future of the Dubai Future Foundation.
In the test, they relied on four deaf people who use the American sign language, they repeated each hand 15 times, so the algorithm distinguished 660 signs, including all letters and numbers from zero to nine.
The university has submitted a patent application for this innovation, and researchers hope to develop the glove and algorithm further, to increase its linguistic inventory and accelerate its translation, to facilitate its marketing and publication.