Date(s) - 19/11/2012
3:30 pm - 4:30 pm
Parsing is a core natural language processing technology, but most parsing work has been developed on written text. Spoken language poses challenges for systems trained on text due to the presence of disfluencies, recognizer errors, and differences associated with word choice and grammatical style. Parsing language models have been shown to be useful for improving speech transcription in terms of reducing word error rate, but we argue that parsing has benefits that go beyond word error rate. With increasing interest in language processing applied to spoken documents, there are now several applications that show a benefit from explicitly parsing speech and using parsing as an objective. This talk illustrates the impact that parsing can have on speech transcription and surveys examples of applications that directly benefit from parsing speech. We also discuss ways in which parsers can be modified to be more effective with spoken language.
Mari Ostendorf is Professor of Electrical Engineering at the University of Washington and Adjunct in Computer Science and Linguistics. Her research interests are in dynamic and linguistically motivated statistical models for speech and language processing that consider the interaction of topic, genre and register. She is a Fellow of the IEEE and ISCA. Prof Ostendorf is a leading researcher in spoken language processing and has made significant (and long-lasting) contributions in speech recognition, speech synthesis, prosodic analysis, and computational linguistics. Her current research interests include: Low resource language modelling; Computational modelling of prosody for spoken document processing; Use of parsing in speech recognition; Language technology for education applications; Extracting social roles and relation information from spoken and written discussions.