Some patients with speech deficits, such as individuals with laryngectomy or brain injury, cannot vocalize. Others, such as Special Forces soldiers, may need to rely on voiceless communication for covert operations or noisy environments. Imagine a wearable technology that gives voice to the voiceless, or provides silent communication.
We are collaborating with researchers from BAE Systems, the Massachusetts General Hospital Voice Center and Northeastern University (Boston, USA) to develop a non-vocal speech communication device that translates facial muscle signals produced when silently mouthing words into automated speech.Current algorithms can process continuous speech from unimpaired and speech dysfunction subjects with less than 10% error rate from a 2,500 word vocabulary. Improvements in the detection technology and in the signal processing algorithms are under development to render a practical implementation with a reduced error rate and for prosodic elements of speech.
Our preliminary results are detailed in Publication list below.
Support
The views expressed in these materials do not necessarily reflect the official policies of the U.S. Department of Defense, U.S. Department of the Interior, U.S. Department of Veterans Affairs, U.S. Department of Health and Human Services, the NIH or its components; nor does the inclusion of trade names/logos/trademarks/or references to outside entities constitute or imply an endorsement by any Federal entity.