Language technology in a predictive, restricted on-screen keyboard with dynamic layout for severely disabled people

Authors:
Anders Sewerin Johansen;John Paulin Hansen;Dan Witzner Hansen;Kenji Itoh;Satoru Mashino
Affiliations:
The IT University of Copenhagen, Copenhagen, Denmark;The IT University of Copenhagen, Copenhagen, Denmark;The IT University of Copenhagen, Copenhagen, Denmark;Tokyo Inst. of Technology, Tokyo, Japan;Tokyo Inst. of Technology, Tokyo, Japan
Venue:
TextEntry '03 Proceedings of the 2003 EACL Workshop on Language Modeling for Text Entry Methods
Year:
2003

Citing 4
Cited 4

The Reactive Keyboard: A Predictive Typing Aid

Computer
The use of eye movements in human-computer interaction techniques: what you look at is what you get

ACM Transactions on Information Systems (TOIS) - Special issue on computer—human interaction
Text input for mobile devices: comparing model prediction to actual performance

Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Twenty years of eye typing: systems and design issues

ETRA '02 Proceedings of the 2002 symposium on Eye tracking research & applications

Keysurf: a character controlled browser for people with physical disabilities

Proceedings of the 17th international conference on World Wide Web
Scanning methods and language modeling for binary switch typing

SLPAT '10 Proceedings of the NAACL HLT 2010 Workshop on Speech and Language Processing for Assistive Technologies
Humsher: a predictive keyboard operated by humming

The proceedings of the 13th international ACM SIGACCESS conference on Computers and accessibility
Performing Locomotion Tasks in Immersive Computer Games with an Adapted Eye-Tracking Interface

ACM Transactions on Accessible Computing (TACCESS)

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes the GazeTalk augmentative and alternative communications (AAC) system, and presents results from two user studies of initial typing rates among novice users. GazeTalk can be operated using eye tracking, mouse or other pointing devices. The system presents the user with a user interface that is based on 12 large on-screen buttons. GazeTalk supports a wide range of configurations, including several variants of probabilistic or ambiguous/clustered keyboards. The language model used in GazeTalk is based on a corpus constructed on the basis of text extracted from Usenet discussion groups. The results from the user studies indicated that the prediction-based input system was less efficient than a static layout. However, user comments suggest that this was mainly caused by design related factors, which were not directly related to the basic design principles. In the next design iteration, we aim to improve the design by eliminating the problems and to increase the quality of the language model by using a significantly larger training corpus.