Wouldn’t it be wonderful if we all learned sign language? Without a doubt, it would be a truly inclusive tool that would rationalize our communication skills. But there is another option. Create an instant translator. That’s what bioengineers from the University of California at Los Angeles (UCLA) thought. They designed the glove that translates sign language. And in real time.
It is a glove-like device. Work through a smartphone application. The research was published in the journal “Nature Electronics”.
Easy communication, easy learning
“We hope this opens up new opportunities. People who use sign language would communicate directly with those who don’t. And without anyone else having to translate for them,” Jun Chen said. He is an assistant professor of bioengineering at Samueli UCLA School of Engineering, also key investigator, “We hope it can help more people learn sign language.”
The system includes a pair of gloves with slim and elastic sensors. They extend along each of the five fingers. These sensors consist of electrically conductive wires. They collect hand movements and finger positions. These represent individual letters, numbers, words and phrases.
The device then converts finger movements into electrical signals. They are supplied on a coin-sized circuit board that is worn around the wrist. The board transmits these signals wirelessly to a smartphone. It translates them into spoken words at a rate of approximately one word per second.
The researchers also added adhesive sensors to the faces of the evaluators. Especially between the eyebrows and on one side of the mouth. They try to grasp the facial expressions that are part of American sign language.
Easy and accessible
Previous portable systems that offered translation into American sign language were limited. The designs were bulky and heavy or difficult to carry, Chen recalls.
The device is made of cheap and light elastic polymers. However, they are durable. Electronic sensors are also very flexible and inexpensive.
When testing the device, the researchers worked with four deaf and American sign language users. Users repeated every hand movement 15 times. A custom machine learning algorithm converted these gestures into letters, numbers, and words. The glove that translates sign language recognized every letter of the alphabet and the numbers from 0 to 9 among 660 other characters. Isn’t it wonderful?