CHI
2008
ACM
14 years 12 months ago
2008
ACM
In a typical speech dictation interface, the recognizer's bestguess is displayed as normal, unannotated text. This ignores potentially useful information about the recognizer...