| J99-4003 | We use two measures to compare | speech repair detection | . The first measure , referred |
| P97-1033 | column gives the results of adding | speech repair detection | to the POS model . The third |
| E97-1033 | column gives the results of adding | speech repair detection | to the POS model . The third |
| J99-4003 | boundaries by 9.1 % . Column four adds | speech repair detection | , which further reduces the error |
| W11-1405 | in sentence units ( SUs ) and | speech repairs detection | ( Liu , 2004 ) . Hence , we use |
| W98-1121 | reduction of 11.2 % . This shows that | speech repair detection | profits from being able to make |
| P97-1033 | . Furthermore , adding in the | speech repair detection | and correction further improves |
| E97-1033 | . Furthermore , adding in the | speech repair detection | and correction further improves |
| W11-1405 | research on sentence boundary and | speech repair detections | ( Shriberg et al. , 2000 ; Shriberg |
| P97-1033 | information , and the fourth column adds | speech repair detection | and correction . We see that |
| E97-1033 | information , and the fourth column adds | speech repair detection | and correction . We see that |
| N09-1074 | speech repair . In that paper | speech repair detection | accuracy was increased by explicitly |
| W98-1121 | Section 3 . The third column adds in | speech repair detection | and correction , boundary tone |
| A97-1010 | Allen ( 1994 ) divide English | speech repairs detection | of the interruption point of |
| J99-4003 | augments the POS-based model with | speech repair detection | and correction , which improves |
