Text entry by gaze is a useful means of hands-free interaction that is applicable in settings where dictation suffers from poor voice recognition or where spoken words and sentences jeopardize privacy or confidentiality. However, text entry by gaze still shows inferior performance and it quickly exhausts its users. We introduce text entry by gaze and nonverbal voice inputs as a hands-free text entry. We review related literature to converge to word-wise text entry by analysis of gaze paths that are temporally constrained by nonverbal voice inputs. We develop and evaluate two design choices. The first requires two voice inputs to indicate the start and end of a word. The second interprets one continuous voice input as indicating the start and end of a word. Our method achieved statistically significant higher text entry rates and more comfortable interaction in an experiment with 12 participants.
Speaker: Ramin Hedeshy