A lexically-driven algorithm for disfluency detection

Authors:
Matthew Snover;Bonnie Dorr;Richard Schwartz
Affiliations:
University of Maryland, MD;University of Maryland, MD;BBN, Columbia, MD
Venue:
HLT-NAACL-Short '04 Proceedings of HLT-NAACL 2004: Short Papers
Year:
2004

Citing 4
Cited 7

Transformation-based error-driven learning and natural language processing: a case study in part-of-speech tagging

Computational Linguistics
Speech repairs, intonational phrases, and discourse markers: modeling speakers' utterances in spoken dialogue

Computational Linguistics
Deterministic parsing of syntactic non-fluencies

ACL '83 Proceedings of the 21st annual meeting on Association for Computational Linguistics
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

Hybrid Multi-step Disfluency Detection

MLMI '08 Proceedings of the 5th international workshop on Machine Learning for Multimodal Interaction
Improved features and models for detecting edit disfluencies in transcribing spontaneous Mandarin speech

IEEE Transactions on Audio, Speech, and Language Processing
Interruption Point Detection of Spontaneous Speech Using Inter-Syllable Boundary-Based Prosodic Features

ACM Transactions on Asian Language Information Processing (TALIP)
Automatic identification of discourse markers in dialogues: An in-depth study of like and well

Computer Speech and Language
The impact of language models and loss functions on repair disfluency detection

HLT '11 Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies - Volume 1
Contextual maximum entropy model for edit disfluency detection of spontaneous speech

ISCSLP'06 Proceedings of the 5th international conference on Chinese Spoken Language Processing
A monotonic statistical machine translation approach to speaking style transformation

Computer Speech and Language

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper describes a transformation-based learning approach to disfluency detection in speech transcripts using primarily lexical features. Our method produces comparable results to two other systems that make heavy use of prosodic features, thus demonstrating that reasonable performance can be achieved without extensive prosodic cues. In addition, we show that it is possible to facilitate the identification of less frequently disfluent discourse markers by taking speaker style into account.