Autism Speaks, Science, Grants Program

Brian Roark received his Ph.D. in Linguistics from Brown University in May of 2001, and then spent three years as a Senior Technical Staff Member at AT&T Labs – Research, in the Speech Algorithms department. In July 2004, he joined the Center for Spoken Language Understanding (CSLU) and the Department of Computer Science & Electrical Engineering at the OGI School of Science & Engineering of the Oregon Health & Science University, as an assistant professor. He is now Associate Professor in CSLU and the newly formed Division of Biomedical Computer Science. His research interests include language modeling for automatic speech recognition (ASR) and assistive and augmentative communication (AAC); supervised and unsupervised model adaptation to novel domains; discriminative language modeling; statistical parsing of text and speech; and spoken language processing for diagnosis of neurodevelopmental and neurodegenerative disorders.

In addition to basic level computer science research, Dr. Roark is currently involved in projects to apply speech and language processing techniques to automate and enhance diagnostic methods for neurodevelopmental disorders such as Autism Spectrum Disorder (ASD), as well as for neurodegenerative conditions such as Mild Cognitive Impairment (MCI). He is also involved in projects to provide brain-computer interface technologies for assistive communication, with a focus that includes both individuals with locked-in syndrome and children with ASD.

Selected publications:
Brian Roark. (In submission). Open vocabulary language modeling for binary response typing interfaces.

Brian Roark and Asaf Bachrach. (In submission). Deriving lexical and syntactic expectation-based measures for psycholinguistic modeling via incremental parsing.

Brian Roark and Richard Sproat. 2007. Computational Approaches to Morphology and Syntax. Oxford University Press, Oxford.

Brian Roark, Margaret Mitchell and Kristy Hollingshead. 2007. Syntactic complexity measures for detecting Mild Cognitive Impairment. In Proceedings of the ACL 2007 Workshop on Biomedical Natural Language Processing (BioNLP), pp. 1-8.

Brian Roark, John-Paul Hosom, Margaret Mitchell and Jeffrey A. Kaye. 2007. Automatically derived spoken language markers for detecting Mild Cognitive Impairment. In Proceedings of the 2nd International Conference on Technology and Aging (ICTA).