HLT '01 Proceedings of the first international conference on Human language technology research
Edit detection and parsing for transcribed speech
NAACL '01 Proceedings of the second meeting of the North American Chapter of the Association for Computational Linguistics on Language technologies
On using written language training data for spoken language modeling
HLT '94 Proceedings of the workshop on Human Language Technology
A TAG-based noisy channel model of speech repairs
ACL '04 Proceedings of the 42nd Annual Meeting on Association for Computational Linguistics
A progressive feature selection algorithm for ultra large feature spaces
ACL-44 Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics
Effective use of prosody in parsing conversational speech
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Maximum expected F-measure training of logistic regression models
HLT '05 Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing
Minimum risk annealing for training log-linear models
COLING-ACL '06 Proceedings of the COLING/ACL on Main conference poster sessions
A lexically-driven algorithm for disfluency detection
HLT-NAACL-Short '04 Proceedings of HLT-NAACL 2004: Short Papers
Broad-coverage parsing using human-like memory constraints
Computational Linguistics
A two-step approach to sentence compression of spoken utterances
ACL '12 Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Short Papers - Volume 2
Linking uncertainty in physicians' narratives to diagnostic correctness
ExProM '12 Proceedings of the Workshop on Extra-Propositional Aspects of Meaning in Computational Linguistics
Hi-index | 0.01 |
Unrehearsed spoken language often contains disfluencies. In order to correctly interpret a spoken utterance, any such disfluencies must be identified and removed or otherwise dealt with. Operating on transcripts of speech which contain disfluencies, we study the effect of language model and loss function on the performance of a linear reranker that rescores the 25-best output of a noisy-channel model. We show that language models trained on large amounts of non-speech data improve performance more than a language model trained on a more modest amount of speech data, and that optimising f-score rather than log loss improves disfluency detection performance. Our approach uses a log-linear reranker, operating on the top n analyses of a noisy channel model. We use large language models, introduce new features into this reranker and examine different optimisation strategies. We obtain a disfluency detection f-scores of 0.838 which improves upon the current state-of-the-art.