Matches in SemOpenAlex for { <https://semopenalex.org/work/W2131385129> ?p ?o ?g. }
Showing items 1 to 86 of
86
with 100 items per page.
- W2131385129 endingPage "955" @default.
- W2131385129 startingPage "945" @default.
- W2131385129 abstract "In this paper, we propose a dynamic in-search data selection method to diagnose competing information automatically from speech data. In our method, the Viterbi beam search is used to decode all training data. During decoding, all partial paths within the beam are examined to identify the so-called competing-token and true-token sets for each individual hidden Markov model (HMM). In this work, the collected data tokens are used for acoustic modeling and utterance verification as two specific examples. In acoustic modeling, the true-token sets are used to adapt HMMs with a sequential maximum a posteriori adaptation method, while a generalized probabilistic descent-based discriminative training method is proposed to improve HMMs based on competing-token sets. In utterance verification, under the framework of likelihood ratio testing, the true-token sets are employed to train positive models for the null hypothesis and the competing-token sets are used to estimate negative models for the alternative hypothesis. All the proposed methods are evaluated in Bell Laboratories communicator system. Experimental results show that the new acoustic modeling method can consistently improve recognition performance over our best maximum likelihood estimation models, roughly 1% absolute reduction in word error rate. The results also show the new verification models can significantly improve the performance of utterance verification over the conventional anti models, almost relatively 30% reduction of equal error rate when identifying misrecognized words from the recognition results." @default.
- W2131385129 created "2016-06-24" @default.
- W2131385129 creator A5008456610 @default.
- W2131385129 creator A5065394791 @default.
- W2131385129 creator A5066868860 @default.
- W2131385129 date "2005-09-01" @default.
- W2131385129 modified "2023-09-23" @default.
- W2131385129 title "A dynamic in-search data selection method with its applications to acoustic modeling and utterance verification" @default.
- W2131385129 cites W155109787 @default.
- W2131385129 cites W1599512239 @default.
- W2131385129 cites W181050861 @default.
- W2131385129 cites W1877570817 @default.
- W2131385129 cites W2003123121 @default.
- W2131385129 cites W2006258746 @default.
- W2131385129 cites W2008826876 @default.
- W2131385129 cites W2034537249 @default.
- W2131385129 cites W2053967299 @default.
- W2131385129 cites W2063541597 @default.
- W2131385129 cites W2069631319 @default.
- W2131385129 cites W2086267546 @default.
- W2131385129 cites W2100969003 @default.
- W2131385129 cites W2123031544 @default.
- W2131385129 cites W2128946969 @default.
- W2131385129 cites W2134659216 @default.
- W2131385129 cites W2135936685 @default.
- W2131385129 cites W2136836257 @default.
- W2131385129 cites W2148428486 @default.
- W2131385129 cites W2158069733 @default.
- W2131385129 cites W2158289097 @default.
- W2131385129 cites W2160246131 @default.
- W2131385129 cites W2158592557 @default.
- W2131385129 doi "https://doi.org/10.1109/tsa.2005.851947" @default.
- W2131385129 hasPublicationYear "2005" @default.
- W2131385129 type Work @default.
- W2131385129 sameAs 2131385129 @default.
- W2131385129 citedByCount "17" @default.
- W2131385129 countsByYear W21313851292012 @default.
- W2131385129 crossrefType "journal-article" @default.
- W2131385129 hasAuthorship W2131385129A5008456610 @default.
- W2131385129 hasAuthorship W2131385129A5065394791 @default.
- W2131385129 hasAuthorship W2131385129A5066868860 @default.
- W2131385129 hasConcept C11413529 @default.
- W2131385129 hasConcept C153180895 @default.
- W2131385129 hasConcept C154945302 @default.
- W2131385129 hasConcept C23224414 @default.
- W2131385129 hasConcept C2775852435 @default.
- W2131385129 hasConcept C28490314 @default.
- W2131385129 hasConcept C38652104 @default.
- W2131385129 hasConcept C40969351 @default.
- W2131385129 hasConcept C41008148 @default.
- W2131385129 hasConcept C48145219 @default.
- W2131385129 hasConcept C57273362 @default.
- W2131385129 hasConcept C97931131 @default.
- W2131385129 hasConceptScore W2131385129C11413529 @default.
- W2131385129 hasConceptScore W2131385129C153180895 @default.
- W2131385129 hasConceptScore W2131385129C154945302 @default.
- W2131385129 hasConceptScore W2131385129C23224414 @default.
- W2131385129 hasConceptScore W2131385129C2775852435 @default.
- W2131385129 hasConceptScore W2131385129C28490314 @default.
- W2131385129 hasConceptScore W2131385129C38652104 @default.
- W2131385129 hasConceptScore W2131385129C40969351 @default.
- W2131385129 hasConceptScore W2131385129C41008148 @default.
- W2131385129 hasConceptScore W2131385129C48145219 @default.
- W2131385129 hasConceptScore W2131385129C57273362 @default.
- W2131385129 hasConceptScore W2131385129C97931131 @default.
- W2131385129 hasIssue "5" @default.
- W2131385129 hasLocation W21313851291 @default.
- W2131385129 hasOpenAccess W2131385129 @default.
- W2131385129 hasPrimaryLocation W21313851291 @default.
- W2131385129 hasRelatedWork W1481751294 @default.
- W2131385129 hasRelatedWork W1565453839 @default.
- W2131385129 hasRelatedWork W1966737826 @default.
- W2131385129 hasRelatedWork W2108840068 @default.
- W2131385129 hasRelatedWork W2124094254 @default.
- W2131385129 hasRelatedWork W2133828439 @default.
- W2131385129 hasRelatedWork W2375198324 @default.
- W2131385129 hasRelatedWork W60887625 @default.
- W2131385129 hasRelatedWork W2105459351 @default.
- W2131385129 hasRelatedWork W2182518606 @default.
- W2131385129 hasVolume "13" @default.
- W2131385129 isParatext "false" @default.
- W2131385129 isRetracted "false" @default.
- W2131385129 magId "2131385129" @default.
- W2131385129 workType "article" @default.