Date of Award

2021

Document Type

Open Access Master's Thesis

Degree Name

Master of Science in Computer Science (MS)

Administrative Home Department

Department of Computer Science

Advisor 1

Keith Vertanen

Committee Member 1

Scott Kuhl

Committee Member 2

Timothy Havens

Committee Member 3

Ricardo Eiris

Abstract

We propose an interface for eyes-free text entry using an ambiguous technique and conduct a preliminary user study. We find that user are able to enter text at 19.09 words per minute (WPM) with a 2.08% character error rate (CER) after eight hours of practice. We explore ways to optimize the ambiguous groupings to reduce the number of disambiguation errors, both with and without familiarity constraints. We find that it is feasible to reduce the number of ambiguous groups from six to four. Finally, we explore a technique for presenting word suggestions to users using simultaneous audio feedback. We find that accuracy is quite poor when the words are played fully simultaneously, but improves when a slight delay is added before each voice.

Share

COinS