This report describes a method for structuring dynamic Bayesian networks so that word and sentence-level models can be constructed from low-level phonetic models. This ability is a fundamental prerequisite for large-scale speech recognition systems, and is well-addressed in the context of hidden Markov models. With dynamic Bayesian networks, however, subword units cannot simply be concatenated together, and an entirely different approach is necessary.
Title
Dynamic Bayesian Networks and the Concatenation Problem in Speech Recognition
Published
1996-12-11
Full Collection Name
Electrical Engineering & Computer Sciences Technical Reports
Other Identifiers
CSD-96-927
Type
Text
Extent
9 p
Archive
The Engineering Library
Usage Statement
Researchers may make free and open use of the UC Berkeley Library’s digitized public domain materials. However, some materials in our online collections may be protected by U.S. copyright law (Title 17, U.S.C.). Use or reproduction of materials protected by copyright beyond that allowed by fair use (Title 17, U.S.C. § 107) requires permission from the copyright owners. The use or reproduction of some materials may also be restricted by terms of University of California gift or purchase agreements, privacy and publicity rights, or trademark law. Responsibility for determining rights status and permissibility of any use or reproduction rests exclusively with the researcher. To learn more or make inquiries, please see our permissions policies (https://www.lib.berkeley.edu/about/permissions-policies).