Cornell University

Fei Wang

Anaïs Rameau

The present disclosure describes approaches to voice restoration using personal devices that detect surface electromyographic (sEMG) signals from articulatory muscles for the recognition of silent speech in a patient (such as a patient with total laryngectomy, on voice rest, etc.). A personal device may comprise a cutaneous sensor unit and a control module that wirelessly transmits signals to a computing device capable of, for example, applying a predictive model to signals to generate text or synthesize speech. Methods and systems for training and applying predictive models, and fabricating personal devices, are disclosed.

Growing community of inventors

Ithaca, NY, United States of America

Fei Wang