Matt Huenerfauth - Featured Faculty 2016
Matt Huenerfauth
Golisano College of Computing and Information Sciences
MATT HUENERFAUTH IS AN ASSOCIATE PROFESSOR IN THE DEPARTMENT OF INFORMATION SCIENCES AND TECHNOLOGIES IN THE GOLISANO COLLEGE OF COMPUTING AND INFORMATION SCIENCES.
His research spans human computer interaction and natural language processing, and he is the editor-in-chief of the leading journal in his field of computer accessibility for people with disabilities. He has secured over $2.5 million in research funding, including a National Science Foundation CAREER Award in 2008.
His laboratory includes both deaf and hearing students; they conduct projects investigating the design of linguistic technologies to bene t people who are deaf or hard-of-hearing.
Animations of American Sign Language (ASL): While many members of the Deaf Community prefer to receive information in the form of ASL, providing ASL on the web has been a challenge because videos of humans are difficult to update and maintain. Huenerfauth's team is developing software to convert an easy-to-update script of an ASL sentence into a computer animation. By modeling the way that humans move during ASL (from motion-capture recordings his team has collected), their technology can produce more realistic animations of ASL, which they evaluate in studies with deaf participants.
Educational Tools for ASL Students: Huenerfauth's lab is creating a tool that would allow students who are learning ASL to practice their signing skills by performing ASL into a Kinect video camera, and the software would automatically provide feedback on their signing, to indicate when they have performed specific linguistic elements or common errors.
Automatic Captioning for Meetings: In collaboration with NTID researchers, Huenerfauth's lab is investigating how automatic speech recognition technology could be used to produce captions automatically for one-on-one or small-group meetings between deaf and hearing participants. His team is examining how to improve the accuracy of these captions and how to indicate which words in the output are more trustworthy.
Matt Huenerfauth
Associate Professor
Golisano College of Computing and Information Sciences