FEATURED LINKS
|
|
||||||||||
|
NewsRelease
Speech recognition guided by prosodic models to be studied in NSF grant NEW HAVEN, CT-- Dr. Louis Goldstein, a Senior Scientist at Haskins Laboratories and a Professor of Linguistics at Yale University, has received an award of $298,000 from the National Science Foundation to study landmark-based robust speech recognition using prosody-guided models of speech variability. This award is part of a larger collaborative project for which the principal investigator is Dr. Carol Espy-Wilson of the University of Maryland, College Park. Despite great strides in the development of automatic speech recognition technology, we do not yet have a system with performance comparable to humans in automatically transcribing unrestricted conversational speech, representing many speakers and dialects, and embedded in adverse acoustic environments. This new approach applies new high-dimensional machine learning techniques, constrained by empirical and theoretical studies of speech production and perception, to learn from data the information structures that human listeners extract from speech. To do this, Goldstein and his colleagues will develop large-vocabulary, psychologically realistic models of speech acoustics, pronunciation variability, prosody, and syntax by deriving knowledge representations that reflect those proposed for human speech production and speech perception.
Haskins Laboratories was founded in 1935 by the late Dr. Caryl P. Haskins. This independent research institute has been in New Haven, Connecticut since 1970 when it formalized affiliations with Yale University and the University of Connecticut. The Laboratories' primary research focus is on the science of the spoken and written word. <click here for NewsRelease in PDF format>
|
|||||||||||
| All text © 1996-2007 Haskins Laboratories For further information, contact webmaster@haskins.yale.edu |
|||||||||||