My research interests include speech recognition, language modeling, and human-computer interaction (HCI).
In particular, I am interested in how multiple input modalities can be combined to efficiently correct speech recognition errors.
My goal is to utilize the rich information available to the recognizer to enable efficient user-guided correction.
I am also interested in improving the performance, usability, and accessibility of text entry interfaces.
What's new:
April, 2013 – The new work I'm doing on mobile text entry for the blind and visually-impaired appeared on the front page of the Montana Standard. Montana Tech student Haythem Memmi is working on the project as part of an undergraduate research project funded by NIH-INBRE.
October, 2012 – Our paper on a phoneme-based predictive communication aid won the ACM SIGACCESS Best Student Paper Award at ASSETS 2012. The lead author is Ha Trinh from the University of Dundee.
May, 2012 – I made a music video out of the awesome work of my freshmen computer science students (see below).
March, 2012 – Our paper on the potential of dwell-free eye-typing received Best Paper Honorable Mention at ETRA 2012.
February, 2012 – Our work on improving conversational language modeling for AAC devices was in an article in the New Scientist.