A team of South Korean researchers have developed a silent speech recognition system that is said to identify words by tracking facial movements. According to a report published in euronews.next, the device intends to assist deaf people, who may find it difficult to interact with others using sign language. It is also expected to aid the military or the police in situations where radio transmission is challenging due to background noise.

Silent speech interfaces use strain sensors to detect a person's movements as they mouth words. The new device can track and translate these facial motions into words using a deep learning algorithm, according to the report. "The strain sensor attached to the face stretches and shrinks according to the skin's stretchiness when a person speaks. And the electric properties of the strain sensors change accordingly," Taemin Kim, of Yonsei University School of Electrical and Electronic Engineering, was quoted as saying.