Matches in SemOpenAlex for { <https://semopenalex.org/work/W2103648052> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W2103648052 abstract "Most current state-of-the-art speech recognition systems use the hidden Markov model (HMM) for modeling the acoustical characteristics of a speech signal. In the first-order HMM, speech data are assumed to be independently and identically distributed (iid), meaning that there is no dependency between neighboring feature vectors. Another assumption is that the current vector depends only on the current HMM state. In practice, however, these assumptions are not true. We describe a hybrid HMM/BN (Bayesian network) acoustic model, where the dependency of the current speech vector on the previous vector and on the previous state is also learned and used in speech recognition. This is possible because the state probability distribution is modeled by a BN. Previous instances of the state and speech feature vector are represented by additional variables of the BN and the probabilistic dependencies between them, and their current instances are learned during training. During recognition, the likelihood of the current feature vector is inferred from the BN where the previous state and previous feature vector are treated as hidden. We have evaluated this hybrid HMM/BN model with our LVCSR system by phoneme recognition and by large-vocabulary continuous word recognition tasks. In both cases, we observed improved performance over the conventional Gaussian mixture HMM." @default.
- W2103648052 created "2016-06-24" @default.
- W2103648052 creator A5020994673 @default.
- W2103648052 creator A5039443541 @default.
- W2103648052 date "2006-10-11" @default.
- W2103648052 modified "2023-09-23" @default.
- W2103648052 title "Modeling Successive Frame Dependencies with Hybrid HMM/BN Acoustic Model" @default.
- W2103648052 cites W125097753 @default.
- W2103648052 cites W146926017 @default.
- W2103648052 cites W1502737068 @default.
- W2103648052 cites W1833981795 @default.
- W2103648052 cites W2020335243 @default.
- W2103648052 cites W2051347452 @default.
- W2103648052 cites W2120082522 @default.
- W2103648052 cites W2121652828 @default.
- W2103648052 cites W2149073513 @default.
- W2103648052 cites W2166388645 @default.
- W2103648052 cites W2171795550 @default.
- W2103648052 cites W2342418827 @default.
- W2103648052 cites W2584852661 @default.
- W2103648052 cites W1564576422 @default.
- W2103648052 doi "https://doi.org/10.1109/icassp.2005.1415210" @default.
- W2103648052 hasPublicationYear "2006" @default.
- W2103648052 type Work @default.
- W2103648052 sameAs 2103648052 @default.
- W2103648052 citedByCount "5" @default.
- W2103648052 crossrefType "proceedings-article" @default.
- W2103648052 hasAuthorship W2103648052A5020994673 @default.
- W2103648052 hasAuthorship W2103648052A5039443541 @default.
- W2103648052 hasConcept C138885662 @default.
- W2103648052 hasConcept C153180895 @default.
- W2103648052 hasConcept C154945302 @default.
- W2103648052 hasConcept C19768560 @default.
- W2103648052 hasConcept C23224414 @default.
- W2103648052 hasConcept C2776401178 @default.
- W2103648052 hasConcept C28490314 @default.
- W2103648052 hasConcept C41008148 @default.
- W2103648052 hasConcept C41895202 @default.
- W2103648052 hasConcept C50644808 @default.
- W2103648052 hasConcept C61224824 @default.
- W2103648052 hasConcept C83665646 @default.
- W2103648052 hasConceptScore W2103648052C138885662 @default.
- W2103648052 hasConceptScore W2103648052C153180895 @default.
- W2103648052 hasConceptScore W2103648052C154945302 @default.
- W2103648052 hasConceptScore W2103648052C19768560 @default.
- W2103648052 hasConceptScore W2103648052C23224414 @default.
- W2103648052 hasConceptScore W2103648052C2776401178 @default.
- W2103648052 hasConceptScore W2103648052C28490314 @default.
- W2103648052 hasConceptScore W2103648052C41008148 @default.
- W2103648052 hasConceptScore W2103648052C41895202 @default.
- W2103648052 hasConceptScore W2103648052C50644808 @default.
- W2103648052 hasConceptScore W2103648052C61224824 @default.
- W2103648052 hasConceptScore W2103648052C83665646 @default.
- W2103648052 hasLocation W21036480521 @default.
- W2103648052 hasOpenAccess W2103648052 @default.
- W2103648052 hasPrimaryLocation W21036480521 @default.
- W2103648052 hasRelatedWork W1493898446 @default.
- W2103648052 hasRelatedWork W1627079870 @default.
- W2103648052 hasRelatedWork W1979449467 @default.
- W2103648052 hasRelatedWork W1996727679 @default.
- W2103648052 hasRelatedWork W1999850543 @default.
- W2103648052 hasRelatedWork W2012687151 @default.
- W2103648052 hasRelatedWork W2085184609 @default.
- W2103648052 hasRelatedWork W2086526471 @default.
- W2103648052 hasRelatedWork W2097963413 @default.
- W2103648052 hasRelatedWork W2139943858 @default.
- W2103648052 hasRelatedWork W2143361917 @default.
- W2103648052 hasRelatedWork W2148487158 @default.
- W2103648052 hasRelatedWork W2149306328 @default.
- W2103648052 hasRelatedWork W2151485937 @default.
- W2103648052 hasRelatedWork W2167712934 @default.
- W2103648052 hasRelatedWork W2169990695 @default.
- W2103648052 hasRelatedWork W2369520313 @default.
- W2103648052 hasRelatedWork W2544397158 @default.
- W2103648052 hasRelatedWork W2751056911 @default.
- W2103648052 hasRelatedWork W3131321118 @default.
- W2103648052 isParatext "false" @default.
- W2103648052 isRetracted "false" @default.
- W2103648052 magId "2103648052" @default.
- W2103648052 workType "article" @default.