B#: chord-based correction for multitouch braille input

Hugo Nicolau; Kyle Montague; Tiago Guerreiro; João Guerreiro; Vicki L. Hanson

doi:10.1145/2556288.2557269

B#: chord-based correction for multitouch braille input

Hugo Nicolau, Kyle Montague, Tiago Guerreiro, João Guerreiro, Vicki L. Hanson

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

22 Citations (Scopus)

Abstract

Braille has paved its way into mobile touchscreen devices, providing faster text input for blind people. This advantage comes at the cost of accuracy, as chord typing over a flat surface has proven to be highly error prone. A misplaced finger on the screen translates into a different or unrecognized character. However, the chord itself gathers information that can be leveraged to improve input performance. We present B#, a novel correction system for multitouch Braille input that uses chords as the atomic unit of information rather than characters. Experimental results on data collected from 11 blind people revealed that B# is effective in correcting errors at character-level, thus providing opportunities for instant corrections of unrecognized chords; and at word-level, where it outperforms a popular spellchecker by providing correct suggestions for 72% of incorrect words (against 38%). We finish with implications for designing chord-based correction system and avenues for future work.

Original language	English
Title of host publication	CHI '14
Subtitle of host publication	Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
Place of Publication	New York
Publisher	Association for Computing Machinery
Pages	1705-1708
Number of pages	4
ISBN (Print)	9781450324731
DOIs	https://doi.org/10.1145/2556288.2557269
Publication status	Published - 26 Apr 2014
Event	32nd Annual ACM CHI Conference on Human Factors in Computing Systems - Metro Toronto Convention Centre, Toronto, Canada Duration: 26 Apr 2014 → 1 May 2014 http://chi2014.acm.org/

Conference

Conference	32nd Annual ACM CHI Conference on Human Factors in Computing Systems
Abbreviated title	CHI 2014
Country/Territory	Canada
City	Toronto
Period	26/04/14 → 1/05/14
Internet address	http://chi2014.acm.org/

Access to Document

10.1145/2556288.2557269

Cite this

@inproceedings{9f976e50cbeb49f2b371485a48d4e8b8,

title = "B\#: chord-based correction for multitouch braille input",

abstract = "Braille has paved its way into mobile touchscreen devices, providing faster text input for blind people. This advantage comes at the cost of accuracy, as chord typing over a flat surface has proven to be highly error prone. A misplaced finger on the screen translates into a different or unrecognized character. However, the chord itself gathers information that can be leveraged to improve input performance. We present B\#, a novel correction system for multitouch Braille input that uses chords as the atomic unit of information rather than characters. Experimental results on data collected from 11 blind people revealed that B\# is effective in correcting errors at character-level, thus providing opportunities for instant corrections of unrecognized chords; and at word-level, where it outperforms a popular spellchecker by providing correct suggestions for 72\% of incorrect words (against 38\%). We finish with implications for designing chord-based correction system and avenues for future work.",

author = "Hugo Nicolau and Kyle Montague and Tiago Guerreiro and Jo{\~a}o Guerreiro and Hanson, \{Vicki L.\}",

year = "2014",

month = apr,

day = "26",

doi = "10.1145/2556288.2557269",

language = "English",

isbn = "9781450324731",

pages = "1705--1708",

booktitle = "CHI '14",

publisher = "Association for Computing Machinery",

note = "32nd Annual ACM CHI Conference on Human Factors in Computing Systems, CHI 2014 ; Conference date: 26-04-2014 Through 01-05-2014",

url = "http://chi2014.acm.org/",

}

Nicolau, H, Montague, K, Guerreiro, T, Guerreiro, J & Hanson, VL 2014, B#: chord-based correction for multitouch braille input. in CHI '14 : Proceedings of the SIGCHI Conference on Human Factors in Computing Systems. Association for Computing Machinery, New York, pp. 1705-1708, 32nd Annual ACM CHI Conference on Human Factors in Computing Systems, Toronto, Canada, 26/04/14. https://doi.org/10.1145/2556288.2557269

TY - GEN

T1 - B#

T2 - 32nd Annual ACM CHI Conference on Human Factors in Computing Systems

AU - Nicolau, Hugo

AU - Montague, Kyle

AU - Guerreiro, Tiago

AU - Guerreiro, João

AU - Hanson, Vicki L.

PY - 2014/4/26

Y1 - 2014/4/26

N2 - Braille has paved its way into mobile touchscreen devices, providing faster text input for blind people. This advantage comes at the cost of accuracy, as chord typing over a flat surface has proven to be highly error prone. A misplaced finger on the screen translates into a different or unrecognized character. However, the chord itself gathers information that can be leveraged to improve input performance. We present B#, a novel correction system for multitouch Braille input that uses chords as the atomic unit of information rather than characters. Experimental results on data collected from 11 blind people revealed that B# is effective in correcting errors at character-level, thus providing opportunities for instant corrections of unrecognized chords; and at word-level, where it outperforms a popular spellchecker by providing correct suggestions for 72% of incorrect words (against 38%). We finish with implications for designing chord-based correction system and avenues for future work.

AB - Braille has paved its way into mobile touchscreen devices, providing faster text input for blind people. This advantage comes at the cost of accuracy, as chord typing over a flat surface has proven to be highly error prone. A misplaced finger on the screen translates into a different or unrecognized character. However, the chord itself gathers information that can be leveraged to improve input performance. We present B#, a novel correction system for multitouch Braille input that uses chords as the atomic unit of information rather than characters. Experimental results on data collected from 11 blind people revealed that B# is effective in correcting errors at character-level, thus providing opportunities for instant corrections of unrecognized chords; and at word-level, where it outperforms a popular spellchecker by providing correct suggestions for 72% of incorrect words (against 38%). We finish with implications for designing chord-based correction system and avenues for future work.

UR - https://www.scopus.com/pages/publications/84900415518

U2 - 10.1145/2556288.2557269

DO - 10.1145/2556288.2557269

M3 - Conference contribution

AN - SCOPUS:84900415518

SN - 9781450324731

SP - 1705

EP - 1708

BT - CHI '14

PB - Association for Computing Machinery

CY - New York

Y2 - 26 April 2014 through 1 May 2014

ER -

B#: chord-based correction for multitouch braille input

Abstract

Conference

Access to Document

Other files and links

Fingerprint