J99-4003 two measures to compare speech repair detection . The first measure , referred
E97-1033 gives the results of adding speech repair detection to the POS model . The third
J99-4003 9.1 % . Column four adds speech repair detection , which further reduces the error
P11-1071 trained on large corpora to increase repair detection accuracy performance . There
D14-1009 This is not measured in standard repair detection on Switchboard . To investigate
H93-1066 recognition systems , enabling repair detection for a majority of repairs using
E97-1033 Furthermore , adding in the speech repair detection and correction further improves
H93-1066 -LSB- 2 \ -RSB- . Hindle decouples repair detection from repair correction . His
E97-1033 the fourth column adds speech repair detection and correction . We see that
N09-1074 repair . In that paper speech repair detection accuracy was increased by explicitly
N06-2019 benefit derived here from oracle repair detection should be realizable in practice
D14-1009 the challenges for incremental repair detection : computational com - plexity
A97-1010 ( 1994 ) divide English speech repairs detection of the interruption point of
N06-2019 be the case . Results shown for repair detection accuracy and its impact on parsing
P93-1007 integration of speech cues into repair detection is that of Hindle ( 1983 ) ,
D14-1009 2013 ) , a useful feature for repair detection ( Lease et al. , 2006 ; Qian
P04-1005 significantly improves the accuracy of repair detection and correction . Second , we
D14-1009 with state-of-the-art incremental repair detection methods , but with better incremental
D14-1009 . <title> Strongly Incremental Repair Detection Systems Faculty of and </title>
J99-4003 the POS-based model with speech repair detection and correction , which improves
hide detail