Paper: A TAG-Based Noisy-Channel Model Of Speech Repairs

ACL ID P04-1005
Title A TAG-Based Noisy-Channel Model Of Speech Repairs
Venue Annual Meeting of the Association of Computational Linguistics
Session Main Conference
Year 2004
Authors

This paper describes a noisy channel model of speech repairs, which can identify and correct repairs in speech transcripts. A syntactic parser is used as the source model, and a novel type of TAG-based transducer is the channel model. The use of TAG is motivated by the intuition that the reparandum is a ough copy" of the repair. The model is trained and tested on the Switchboard dis uency-annotated corpus.