TY - GEN
T1 - Effects of language modeling and its personalization on touchscreen typing performance
AU - Fowler, Andrew
AU - Partridge, Kurt
AU - Chelba, Ciprian
AU - Bi, Xiaojun
AU - Ouyang, Tom
AU - Zhai, Shumin
N1 - Publisher Copyright: © Copyright 2015 ACM.
PY - 2015/4/18
Y1 - 2015/4/18
N2 - Modern smartphones correct typing errors and learn userspecific words (such as proper names). Both techniques are useful, yet little has been published about their technical specifics and concrete benefits. One reason is that typing accuracy is difficult to measure empirically on a large scale. We describe a closed-loop, smart touch keyboard (STK) evaluation system that we have implemented to solve this problem. It includes a principled typing simulator for generating human-like noisy touch input, a simple-yet-effective decoder for reconstructing typed words from such spatial data, a large web-scale background language model (LM), and a method for incorporating LM personalization. Using the Enron email corpus as a personalization test set, we show for the first time at this scale that a combined spatial/language model reduces word error rate from a pre-model baseline of 38.4% down to 5.7%, and that LM personalization can improve this further to 4.6%.
AB - Modern smartphones correct typing errors and learn userspecific words (such as proper names). Both techniques are useful, yet little has been published about their technical specifics and concrete benefits. One reason is that typing accuracy is difficult to measure empirically on a large scale. We describe a closed-loop, smart touch keyboard (STK) evaluation system that we have implemented to solve this problem. It includes a principled typing simulator for generating human-like noisy touch input, a simple-yet-effective decoder for reconstructing typed words from such spatial data, a large web-scale background language model (LM), and a method for incorporating LM personalization. Using the Enron email corpus as a personalization test set, we show for the first time at this scale that a combined spatial/language model reduces word error rate from a pre-model baseline of 38.4% down to 5.7%, and that LM personalization can improve this further to 4.6%.
KW - Keyboard error correction
KW - Language modeling
KW - Mobile text entry
UR - https://www.scopus.com/pages/publications/84951023277
U2 - 10.1145/2702123.2702503
DO - 10.1145/2702123.2702503
M3 - Conference contribution
T3 - Conference on Human Factors in Computing Systems - Proceedings
SP - 649
EP - 658
BT - CHI 2015 - Proceedings of the 33rd Annual CHI Conference on Human Factors in Computing Systems
PB - Association for Computing Machinery
T2 - 33rd Annual CHI Conference on Human Factors in Computing Systems, CHI 2015
Y2 - 18 April 2015 through 23 April 2015
ER -