A Syntactic Time-Series Model for Parsing Fluent and Disfluent Speech

This paper describes an incremental approach to parsing transcribed spontaneous speech containing disfluencies with a Hierarchical Hidden Markov Model (HHMM). This model makes use of the right-corner transform, which has been shown to increase non-incremental parsing accuracy on transcribed spontaneous speech (Miller and Schuler, 2008), using trees transformed in this manner to train the HHMM parser. Not only do the representations used in this model align with structure in speech repairs, but as an HMM-like time-series model, it can be directly integrated into conventional speech recognition systems run on continuous streams of audio. A system implementing this model is evaluated on the standard task of parsing the Switchboard corpus, and achieves an improvement over the standard baseline probabilistic CYK parser.

• Type: Conference
• Year: 2008
• Where: COLING
• Authors: Tim Miller, William Schuler
• Keywords: COLING 2008, Computational Linguistics, Hierarchical Hidden Markov Model, Speech Containing Disfluencies, Spontaneous Speech