Matches in SemOpenAlex for { <https://semopenalex.org/work/W2007243334> ?p ?o ?g. }
- W2007243334 endingPage "1402" @default.
- W2007243334 startingPage "1391" @default.
- W2007243334 abstract "This paper presents a study of two acoustic speaker adaptation techniques applied in the context of the subspace Gaussian mixture model (SGMM) for automatic speech recognition (ASR). First, a model space linear regression based approach is presented for adaptation of SGMM state projection vectors and is referred to as subspace vector adaptation (SVA). Second, an easy to implement realization of constrained maximum likelihood linear regression (CMLLR) is presented for feature space adaptation in the SGMM. Numerically stable procedures for row-by-row estimation of the regression based transformation matrices are presented for both SVA and CMLLR adaptation. These approaches are applied to SGMM models that are estimated using speaker adaptive training (SAT), a technique for estimating more compact speaker independent acoustic models. Unsupervised speaker adaptation performance is evaluated on conversational and read speech task domains and compared to unsupervised adaptation performance obtained using the hidden Markov model-Gaussian mixture model (HMM-GMM) in ASR. It is shown that the feature space and model space adaptation approaches applied to the SGMM provide complementary reductions in word error rate (WER) and provide lower WERs than that obtained using CMLLR adaptation for the HMM-GMM." @default.
- W2007243334 created "2016-06-24" @default.
- W2007243334 creator A5017799984 @default.
- W2007243334 creator A5077077136 @default.
- W2007243334 date "2014-09-01" @default.
- W2007243334 modified "2023-10-18" @default.
- W2007243334 title "Linear Regression Based Acoustic Adaptation for the Subspace Gaussian Mixture Model" @default.
- W2007243334 cites W127709214 @default.
- W2007243334 cites W1599512239 @default.
- W2007243334 cites W1831910457 @default.
- W2007243334 cites W1932968309 @default.
- W2007243334 cites W1981706894 @default.
- W2007243334 cites W2002342963 @default.
- W2007243334 cites W2037740282 @default.
- W2007243334 cites W2054145113 @default.
- W2007243334 cites W2084215616 @default.
- W2007243334 cites W2099826692 @default.
- W2007243334 cites W2104924374 @default.
- W2007243334 cites W2107687614 @default.
- W2007243334 cites W2121981798 @default.
- W2007243334 cites W2123798005 @default.
- W2007243334 cites W2134260474 @default.
- W2007243334 cites W2138632432 @default.
- W2007243334 cites W2145691584 @default.
- W2007243334 cites W2146871184 @default.
- W2007243334 cites W2151402979 @default.
- W2007243334 cites W58456301 @default.
- W2007243334 cites W72757270 @default.
- W2007243334 doi "https://doi.org/10.1109/taslp.2014.2332043" @default.
- W2007243334 hasPublicationYear "2014" @default.
- W2007243334 type Work @default.
- W2007243334 sameAs 2007243334 @default.
- W2007243334 citedByCount "0" @default.
- W2007243334 crossrefType "journal-article" @default.
- W2007243334 hasAuthorship W2007243334A5017799984 @default.
- W2007243334 hasAuthorship W2007243334A5077077136 @default.
- W2007243334 hasConcept C11413529 @default.
- W2007243334 hasConcept C120665830 @default.
- W2007243334 hasConcept C121332964 @default.
- W2007243334 hasConcept C12362212 @default.
- W2007243334 hasConcept C139807058 @default.
- W2007243334 hasConcept C151730666 @default.
- W2007243334 hasConcept C153180895 @default.
- W2007243334 hasConcept C154945302 @default.
- W2007243334 hasConcept C163716315 @default.
- W2007243334 hasConcept C23224414 @default.
- W2007243334 hasConcept C2524010 @default.
- W2007243334 hasConcept C2779343474 @default.
- W2007243334 hasConcept C28490314 @default.
- W2007243334 hasConcept C32834561 @default.
- W2007243334 hasConcept C33923547 @default.
- W2007243334 hasConcept C41008148 @default.
- W2007243334 hasConcept C57493831 @default.
- W2007243334 hasConcept C61224824 @default.
- W2007243334 hasConcept C61326573 @default.
- W2007243334 hasConcept C62520636 @default.
- W2007243334 hasConcept C83665646 @default.
- W2007243334 hasConcept C86803240 @default.
- W2007243334 hasConceptScore W2007243334C11413529 @default.
- W2007243334 hasConceptScore W2007243334C120665830 @default.
- W2007243334 hasConceptScore W2007243334C121332964 @default.
- W2007243334 hasConceptScore W2007243334C12362212 @default.
- W2007243334 hasConceptScore W2007243334C139807058 @default.
- W2007243334 hasConceptScore W2007243334C151730666 @default.
- W2007243334 hasConceptScore W2007243334C153180895 @default.
- W2007243334 hasConceptScore W2007243334C154945302 @default.
- W2007243334 hasConceptScore W2007243334C163716315 @default.
- W2007243334 hasConceptScore W2007243334C23224414 @default.
- W2007243334 hasConceptScore W2007243334C2524010 @default.
- W2007243334 hasConceptScore W2007243334C2779343474 @default.
- W2007243334 hasConceptScore W2007243334C28490314 @default.
- W2007243334 hasConceptScore W2007243334C32834561 @default.
- W2007243334 hasConceptScore W2007243334C33923547 @default.
- W2007243334 hasConceptScore W2007243334C41008148 @default.
- W2007243334 hasConceptScore W2007243334C57493831 @default.
- W2007243334 hasConceptScore W2007243334C61224824 @default.
- W2007243334 hasConceptScore W2007243334C61326573 @default.
- W2007243334 hasConceptScore W2007243334C62520636 @default.
- W2007243334 hasConceptScore W2007243334C83665646 @default.
- W2007243334 hasConceptScore W2007243334C86803240 @default.
- W2007243334 hasIssue "9" @default.
- W2007243334 hasLocation W20072433341 @default.
- W2007243334 hasOpenAccess W2007243334 @default.
- W2007243334 hasPrimaryLocation W20072433341 @default.
- W2007243334 hasRelatedWork W1579116662 @default.
- W2007243334 hasRelatedWork W2053673812 @default.
- W2007243334 hasRelatedWork W2098825707 @default.
- W2007243334 hasRelatedWork W2108334821 @default.
- W2007243334 hasRelatedWork W2123876776 @default.
- W2007243334 hasRelatedWork W2172153733 @default.
- W2007243334 hasRelatedWork W2547150524 @default.
- W2007243334 hasRelatedWork W4285337867 @default.
- W2007243334 hasRelatedWork W919364087 @default.
- W2007243334 hasRelatedWork W35176419 @default.
- W2007243334 hasVolume "22" @default.
- W2007243334 isParatext "false" @default.
- W2007243334 isRetracted "false" @default.
- W2007243334 magId "2007243334" @default.