Show simple item record

dc.contributor.authorRenals, Steve
dc.contributor.authorHochberg, Mike
dc.coverage.spatial4en
dc.date.accessioned2006-05-15T16:05:47Z
dc.date.available2006-05-15T16:05:47Z
dc.date.issued1996-05
dc.identifier.citationAcoustics, Speech, and Signal Processing, 1996. ICASSP-96. Conference Proceedings., 1996 IEEE International Conference on, Volume 1, 7-10 May 1996 Page(s):149 - 152.en
dc.identifier.otherDigital Object Identifier 10.1109/ICASSP.1996.540312
dc.identifier.urihttp://ieeexplore.ieee.org/servlet/opac?punumber=3856
dc.identifier.urihttp://hdl.handle.net/1842/1083
dc.description.abstractThis article further develops and analyses the large vocabulary continuous speech recognition (LVCSR) search strategy reported by Renals and Hochberg (see Proc. ICASSP '95, p.596-9, 1995). In particular, the posterior-based phone deactivation pruning approach has been extended to include phone-dependent thresholds and an improved estimate of the least upper bound on the utterance log-probability has been developed. Analysis of the pruning procedures and of the search's interaction with the language model has also been performed. Experiments were carried out using the ARPA North American Business News task with a 20,000 word vocabulary and a trigram language model. As a result of these improvements and analyses, the computational cost of the recognition process performed by the NOWAY decoder has been substantially reduced.en
dc.format.extent444705 bytes
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherIEEE Signal Processing Societyen
dc.titleEfficient evaluation of the LVCSR search space using the NOWAY decoderen
dc.typeConference Paperen


Files in this item

This item appears in the following Collection(s)

Show simple item record