Text & Speech: Processing Speech by Deriving and Exploiting Linguistic Subcultures
Spoken language understanding systems are error-prone for several reasons, including individual speech variability. This is manifested in many ways, among which are differences in pronunciation, lexical inventory, grammar and disfluencies. There is, however, a lot of evidence pointing to stable language usage within subgroups of a language population. We call these subgroups /linguistic subcultures/. We discuss how to identify linguistic subcultures and how effective it is when applied to tasks in speech and language processing, specifically: a) automatic speech recognition and b) cue phrase identification.
David Guy Brizan is an Assistant Professor in the Department of Computer Science at the University of San Francisco, where he has been since 2016. In that position, he is a co-director of the Machine
Learning, AI, Game Intelligence and Computing at Scale (MAGICS) Lab. Prior to that, he held several positions in public service (the Department of Correction, City of New York) and private companies,
including IBM and United HealthCare.
David completed his Ph.D. at the Graduate Center, City University of New York. His research interests are in speech and natural language processing (NLP), machine learning and databases. Much of his work has
involved processing and deriving speech communities within a language.