NSF Awards Jason Bishop Over $400K for a Giant Speech Database

August 2, 2023

The public database will allow researchers to study differences in people’s speech patterns and what causes them.

Lab equipment:mike and computer
(Credit: Getty Images)

The National Science Foundation has bestowed a generous grant to a project, led by a Graduate Center linguist, to build a speech database.

The funding supports the work of Professor Jason Bishop (GC/College of Staten Island, Linguistics, Speech-Language-Hearing Sciences), a cognitive scientist who specializes in speech prosody — the patterns of rhythm, stress, and intonation of speech.

Learn More About the Ph.D. Program in Linguistics

Jason Bishop headshot
Jason Bishop

The grant provides $438,275 in funding for Bishop and partners, Jelena Krivokapić at the University of Michigan and Laurel MacKenzie at New York University, to construct a speech database to catalog differences in speech production. The publicly available corpus will include recordings of tens of thousands of words by hundreds of native English speakers as well as measurements of the speakers’ cognitive skills, including memory and attention, and personality characteristics, such as empathy and autistic traits. The corpus will allow researchers to explore variations in people’s speech patterns and what causes them — knowledge that could lead to more accurate diagnoses of language disorders and improved speech recognition technology.

Data for the three-year project will be collected at the College of Staten Island. The project is expected to employ about a dozen CUNY students, including several from the Graduate Center.

Learn More About the Ph.D. Program in Speech-Language-Hearing Sciences

Bishop, who directs the CUNY Prosody Laboratory, is interested in understanding how speech sounds are articulated and how they are represented in the minds of speakers and listeners. In a recent paper, published in the Journal of Autism and Developmental Disorders, he and his team examined the effects of higher autistic trait loads on speech intelligibility. His current work looks at sentence processing and individual differences in memory and attention.

The speech corpus is a joint project with New York University and the University of Michigan.

Published by the Office of Communications and Marketing