Show simple item record

dc.contributor.advisorHansen, Eric A.
dc.contributor.authorParihar, Naveen
dc.date2009
dc.date.accessioned2020-05-07T17:44:15Z
dc.date.available2020-05-07T17:44:15Z
dc.identifier.urihttps://hdl.handle.net/11668/17071
dc.description.abstractState-of-the-art speech-recognition systems can successfully perform simple tasks in real-time on most computers, when the tasks are performed in controlled and noise-free environments. However, current algorithms and processors are not yet powerful enough for real-time large-vocabulary conversational speech recognition in noisy, real-world environments. Parallel processing can improve the real-time performance of speech recognition systems and increase their applicability, and developing an effective approach to parallelization is especially important given the recent trend toward multi-core processor design. In this dissertation, we introduce methods for parallelizing a single-pass across-word n-gram lexical-tree based Viterbi recognizer, which is the most popular architecture for Viterbi-based large vocabulary continuous speech recognition. We parallelize two different open-source implementations of such a recognizer, one developed at Mississippi State University and the other developed at Rheinisch-Westfalische Technische Hochschule University in Germany. We describe three methods for parallelization. The first, called parallel fast likelihood computation, parallelizes likelihood computations by decomposing mixtures among CPU cores, so that each core computes the likelihood of the set of mixtures allocated to it. A second method, lexical-tree division, parallelizes the search management component of a speech recognizer by dividing the lexical tree among the cores. A third and alternative method for parallelizing the search-management component of a speech recognizer, called lexical-tree copies decomposition, dynamically distributes the active lexical-tree copies among the cores. All parallelization methods were tested on two and four cores of an Intel Core2 Quad processor and significantly improved real-time performance. Several challenges for parallelizing a lexical-tree based Viterbi speech recognizer are also identified and discussed.
dc.publisherMississippi State University
dc.subject.lcshAutomatic speech recognition.
dc.subject.lcshSpeech processing systems.
dc.subject.lcshHidden Markov models.
dc.subject.lcshParallel algorithms.
dc.subject.otherfast gaussian calculations
dc.subject.otherfast likelihood computations
dc.subject.otherprefix tree
dc.subject.otherlexical tree
dc.subject.otherparallel speech decoding
dc.subject.otherparallel speech recognition
dc.subject.othermulti-core processors
dc.titleParallel Viterbi Search For Continuous Speech Recognition On A Multi-Core Architecture
dc.typeDissertation
dc.publisher.departmentDepartment of Electrical and Computer Engineering.
dc.subject.degreeDoctor of Philosophy
dc.subject.majorComputer Engineering
dc.contributor.committeeLuke, Edward A.
dc.contributor.committeeFowler, James E.
dc.contributor.committeeYounan, Nicholas H.
dc.contributor.committeeBridges, Susan M.


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record