Matches in SemOpenAlex for { <https://semopenalex.org/work/W1823291590> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W1823291590 abstract "Presents three forms of linear transform based speaker adaptation that can give better performance than standard maximum likelihood linear regression (MLLR) adaptation. For unsupervised adaptation, a lattice-based technique is introduced which is compared to MLLR using confidence scores. For supervised adaptation, estimation of the adaptation matrices using the maximum mutual information criterion is discussed which leads to the MMILR approach. Recognition experiments show that lattice MLLR can reduce word error rates on a Switchboard task by 1.4% absolute. For recognition of non-native speech from the Wall Street Journal database, a reduction in word error rate of 10-16% relative was obtained using MMILR compared to standard MLLR." @default.
- W1823291590 created "2016-06-24" @default.
- W1823291590 creator A5002191410 @default.
- W1823291590 creator A5006579215 @default.
- W1823291590 date "2002-11-13" @default.
- W1823291590 modified "2023-09-25" @default.
- W1823291590 title "Improvements in linear transform based speaker adaptation" @default.
- W1823291590 cites W130324013 @default.
- W1823291590 cites W1481751294 @default.
- W1823291590 cites W1651108041 @default.
- W1823291590 cites W1892722218 @default.
- W1823291590 cites W2002342963 @default.
- W1823291590 cites W2037740282 @default.
- W1823291590 cites W2100969003 @default.
- W1823291590 cites W2113094193 @default.
- W1823291590 cites W2129334286 @default.
- W1823291590 cites W2146871184 @default.
- W1823291590 cites W2152439927 @default.
- W1823291590 cites W2163105471 @default.
- W1823291590 cites W70088150 @default.
- W1823291590 cites W94829088 @default.
- W1823291590 doi "https://doi.org/10.1109/icassp.2001.940764" @default.
- W1823291590 hasPublicationYear "2002" @default.
- W1823291590 type Work @default.
- W1823291590 sameAs 1823291590 @default.
- W1823291590 citedByCount "43" @default.
- W1823291590 countsByYear W18232915902013 @default.
- W1823291590 countsByYear W18232915902016 @default.
- W1823291590 countsByYear W18232915902017 @default.
- W1823291590 countsByYear W18232915902020 @default.
- W1823291590 crossrefType "proceedings-article" @default.
- W1823291590 hasAuthorship W1823291590A5002191410 @default.
- W1823291590 hasAuthorship W1823291590A5006579215 @default.
- W1823291590 hasConcept C119857082 @default.
- W1823291590 hasConcept C120665830 @default.
- W1823291590 hasConcept C121332964 @default.
- W1823291590 hasConcept C133892786 @default.
- W1823291590 hasConcept C139807058 @default.
- W1823291590 hasConcept C153180895 @default.
- W1823291590 hasConcept C154945302 @default.
- W1823291590 hasConcept C28490314 @default.
- W1823291590 hasConcept C40969351 @default.
- W1823291590 hasConcept C41008148 @default.
- W1823291590 hasConcept C48921125 @default.
- W1823291590 hasConceptScore W1823291590C119857082 @default.
- W1823291590 hasConceptScore W1823291590C120665830 @default.
- W1823291590 hasConceptScore W1823291590C121332964 @default.
- W1823291590 hasConceptScore W1823291590C133892786 @default.
- W1823291590 hasConceptScore W1823291590C139807058 @default.
- W1823291590 hasConceptScore W1823291590C153180895 @default.
- W1823291590 hasConceptScore W1823291590C154945302 @default.
- W1823291590 hasConceptScore W1823291590C28490314 @default.
- W1823291590 hasConceptScore W1823291590C40969351 @default.
- W1823291590 hasConceptScore W1823291590C41008148 @default.
- W1823291590 hasConceptScore W1823291590C48921125 @default.
- W1823291590 hasLocation W18232915901 @default.
- W1823291590 hasOpenAccess W1823291590 @default.
- W1823291590 hasPrimaryLocation W18232915901 @default.
- W1823291590 hasRelatedWork W130324013 @default.
- W1823291590 hasRelatedWork W1493163583 @default.
- W1823291590 hasRelatedWork W1599512239 @default.
- W1823291590 hasRelatedWork W181370797 @default.
- W1823291590 hasRelatedWork W2002342963 @default.
- W1823291590 hasRelatedWork W2003123121 @default.
- W1823291590 hasRelatedWork W2033565080 @default.
- W1823291590 hasRelatedWork W2037740282 @default.
- W1823291590 hasRelatedWork W2063541597 @default.
- W1823291590 hasRelatedWork W2064283427 @default.
- W1823291590 hasRelatedWork W2097003869 @default.
- W1823291590 hasRelatedWork W2100969003 @default.
- W1823291590 hasRelatedWork W2106554350 @default.
- W1823291590 hasRelatedWork W2121981798 @default.
- W1823291590 hasRelatedWork W2123649031 @default.
- W1823291590 hasRelatedWork W2126033156 @default.
- W1823291590 hasRelatedWork W2146871184 @default.
- W1823291590 hasRelatedWork W2147166770 @default.
- W1823291590 hasRelatedWork W269934362 @default.
- W1823291590 hasRelatedWork W653761051 @default.
- W1823291590 isParatext "false" @default.
- W1823291590 isRetracted "false" @default.
- W1823291590 magId "1823291590" @default.
- W1823291590 workType "article" @default.