Matches in SemOpenAlex for { <https://semopenalex.org/work/W2116247168> ?p ?o ?g. }
Showing items 1 to 59 of
59
with 100 items per page.
- W2116247168 abstract "Maximum-Likelihod Linear Regression (MLLR) transform coefficients have shown to be useful features for text-independent speaker recognition systems. These use MLLR coefficients computed on a Large Vocabulary Continuous Speech Recognition System (LVCSR) as features and Support Vector machines(SVM) classification. However, performance is limited by transcripts, which are often erroneous with high word error rates (WER) for spontaneous telephone speech applications. In this paper, we propose using lattice-based MLLR to overcome this issue. Using wordlattices instead of 1-best hypotheses, more hypotheses can be considered for MLLR estimation and, thus, better models are more likely to be used. As opposed to standard MLLR, language model probabilities are taken into account as well. We show how systems using lattice MLLR outperform standard MLLR systems in the Speaker Recognition Evaluation (SRE) 2006. Comparison to other standard acoustic systems is provided as well." @default.
- W2116247168 created "2016-06-24" @default.
- W2116247168 creator A5027284390 @default.
- W2116247168 creator A5060766811 @default.
- W2116247168 creator A5072713669 @default.
- W2116247168 date "2009-01-01" @default.
- W2116247168 modified "2023-10-18" @default.
- W2116247168 title "Lattice-based MLLR for speaker recognition" @default.
- W2116247168 cites W130324013 @default.
- W2116247168 cites W1553271421 @default.
- W2116247168 cites W160722039 @default.
- W2116247168 cites W1823291590 @default.
- W2116247168 cites W2005422315 @default.
- W2116247168 cites W2105667909 @default.
- W2116247168 cites W2121981798 @default.
- W2116247168 cites W2141279928 @default.
- W2116247168 cites W2146871184 @default.
- W2116247168 cites W2147147599 @default.
- W2116247168 cites W2406690299 @default.
- W2116247168 cites W3145606929 @default.
- W2116247168 cites W32979931 @default.
- W2116247168 cites W6845005 @default.
- W2116247168 hasPublicationYear "2009" @default.
- W2116247168 type Work @default.
- W2116247168 sameAs 2116247168 @default.
- W2116247168 citedByCount "5" @default.
- W2116247168 countsByYear W21162471682013 @default.
- W2116247168 countsByYear W21162471682016 @default.
- W2116247168 crossrefType "proceedings-article" @default.
- W2116247168 hasAuthorship W2116247168A5027284390 @default.
- W2116247168 hasAuthorship W2116247168A5060766811 @default.
- W2116247168 hasAuthorship W2116247168A5072713669 @default.
- W2116247168 hasConcept C133892786 @default.
- W2116247168 hasConcept C153180895 @default.
- W2116247168 hasConcept C154945302 @default.
- W2116247168 hasConcept C28490314 @default.
- W2116247168 hasConcept C41008148 @default.
- W2116247168 hasConceptScore W2116247168C133892786 @default.
- W2116247168 hasConceptScore W2116247168C153180895 @default.
- W2116247168 hasConceptScore W2116247168C154945302 @default.
- W2116247168 hasConceptScore W2116247168C28490314 @default.
- W2116247168 hasConceptScore W2116247168C41008148 @default.
- W2116247168 hasLocation W21162471681 @default.
- W2116247168 hasOpenAccess W2116247168 @default.
- W2116247168 hasPrimaryLocation W21162471681 @default.
- W2116247168 hasRelatedWork W1491159402 @default.
- W2116247168 hasRelatedWork W1813780412 @default.
- W2116247168 hasRelatedWork W2029134149 @default.
- W2116247168 hasRelatedWork W2033914206 @default.
- W2116247168 hasRelatedWork W2042327336 @default.
- W2116247168 hasRelatedWork W2499802997 @default.
- W2116247168 hasRelatedWork W3162054169 @default.
- W2116247168 hasRelatedWork W4297807400 @default.
- W2116247168 hasRelatedWork W4313854686 @default.
- W2116247168 hasRelatedWork W289407349 @default.
- W2116247168 isParatext "false" @default.
- W2116247168 isRetracted "false" @default.
- W2116247168 magId "2116247168" @default.
- W2116247168 workType "article" @default.