| H93-1066 | edit signal site is crucial for | repair correction | . 3 . THE REPAIR INTERVAL MODEL |
| W08-0110 | referring , connective alternative and | repair correction | , had very few instances . The |
| J99-4003 | experiments that include speech | repair correction | , we further relax this rule |
| P93-1007 | repair detection is decoupled from | repair correction | , which requires only that the |
| J99-4003 | . The third column adds speech | repair correction | , which improves the error rate |
| J99-4003 | 3.5 % . Column five adds speech | repair correction | . Curiously , this actually slightly |
| W98-1121 | recognition problem ( without speech | repair correction | and silence information ) . The |
| H93-1066 | decouples repair detection from | repair correction | . His correction strategies rely |
| J99-4003 | intonational phrase detection , | repair correction | , and silence information , and |
| J99-4003 | benefits from modeling speech | repair correction | , intonational phrases , silences |
| J99-4003 | for models that include speech | repair correction | , if a misidentified repair is |
| J99-4003 | point for incorporating speech | repair correction | into a statistical language model |
| W98-1121 | accounts for interactions with speech | repair correction | and intonational phrasing , and |
| P97-1033 | The fourth column adds in speech | repair correction | , and shows that taking into |
| E97-1033 | The fourth column adds in speech | repair correction | , and shows that taking into |
| P93-1007 | intonational cue that may be of use for | repair correction | , namely the prosodic phrasing |
