Researchers from Cornell University in the US have developed EchoSpeech glasses, which recognize soundless speech – lip and mouth movements. To do this, they use acoustic sensors and artificial intelligence.
The EchoSpeech interface requires only a few minutes of training to recognize a person’s commands and transmit them to a smartphone.
Tiny microphones and speakers pick up audio signals on the face and ‘sense’ mouth movements, while the algorithm analyzes them in real-time with an accuracy of about 95%.
Ruidong Zhang, a PhD candidate in information science and author of the eponymous technology book, adds that in the future, EchoSpeech will help people with mutism “regain their voices”.
Currently, EchoSpeech can be used for communication via a smartphone in places where speaking out loud is impossible or inappropriate, such as on crowded transportation or in a library. The glasses have a minimalist design, so there should be no problems with wearing them.