Ruidong Zhang modelling his EchoSpeech spectacles

US doctoral student Ruidong Zhang has developed spectacle frames capable of reading lip and mouth movements to infer speech, providing a hands- and eyes-free input method for a variety of applications.

Based at Cornell University’s Smart Computer Interfaces for Future Interactions (SciFi) Lab, the EchoSpeech spectacles use acoustic sensing and AI to continuously recognise up to 31 unvocalised commands, said Zhang. Traditional speech recognition requires the user to speak aloud, which is frequently inconvenient or socially inappropriate, he explained. “For people who cannot vocalise sound, this silent speech technology could be an excellent input for a voice synthesiser. It could give patients their voices back.”

Zhang said EchoSpeech could be used to silently communicate with others via smartphone or be paired with a stylus and used with design software, potentially eliminating the need for a keyboard and a mouse.