Using Sign Language Recognition for Useful, and Usable, Interfaces for the Deaf Community

Hariri Institute Distinguished Seminar: Co-Sponsored by BU Wheelock Deaf Center

Speaker: Thad Starner, Professor of Computing

Many machine learning researchers believe sign language recognition (SLR) is a solved problem, or is solvable using current techniques. In truth, SLR is still at the stage that speech recognition was in 1990. Similar to efforts in that era, we are gathering datasets to address progressively more difficult SLR problems that are important to the Deaf community. We are also investigating captioning head worn displays to help people who are hard of hearing to better communicate.