Matches in SemOpenAlex for { <https://semopenalex.org/work/W2798928049> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W2798928049 abstract "The front-end factor analysis (FEFA), an extension of principal component analysis (PPCA) tailored to be used with Gaussian mixture models (GMMs), is currently the prevalent approach to extract compact utterance-level features (i-vectors) for automatic speaker verification (ASV) systems. Little research has been conducted comparing FEFA to the conventional PPCA applied to maximum a posteriori (MAP) adapted GMM supervectors. We study several alternative methods, including PPCA, factor analysis (FA), and two supervised approaches, supervised PPCA (SPPCA) and the recently proposed probabilistic partial least squares (PPLS), to compress MAP-adapted GMM supervectors. The resulting i-vectors are used in ASV tasks with a probabilistic linear discriminant analysis (PLDA) back-end. We experiment on two different datasets, on the telephone condition of NIST SRE 2010 and on the recent VoxCeleb corpus collected from YouTube videos containing celebrity interviews recorded in various acoustical and technical conditions. The results suggest that, in terms of ASV accuracy, the supervector compression approaches are on a par with FEFA. The supervised approaches did not result in improved performance. In comparison to FEFA, we obtained more than hundred-fold (100x) speedups in the total variability model (TVM) training using the PPCA and FA supervector compression approaches." @default.
- W2798928049 created "2018-05-07" @default.
- W2798928049 creator A5031094144 @default.
- W2798928049 creator A5043168931 @default.
- W2798928049 date "2018-05-03" @default.
- W2798928049 modified "2023-09-27" @default.
- W2798928049 title "Supervector Compression Strategies to Speed up I-Vector System Development" @default.
- W2798928049 hasPublicationYear "2018" @default.
- W2798928049 type Work @default.
- W2798928049 sameAs 2798928049 @default.
- W2798928049 citedByCount "0" @default.
- W2798928049 crossrefType "posted-content" @default.
- W2798928049 hasAuthorship W2798928049A5031094144 @default.
- W2798928049 hasAuthorship W2798928049A5043168931 @default.
- W2798928049 hasConcept C105795698 @default.
- W2798928049 hasConcept C111219384 @default.
- W2798928049 hasConcept C111472728 @default.
- W2798928049 hasConcept C138885662 @default.
- W2798928049 hasConcept C153180895 @default.
- W2798928049 hasConcept C154945302 @default.
- W2798928049 hasConcept C27438332 @default.
- W2798928049 hasConcept C2775852435 @default.
- W2798928049 hasConcept C28490314 @default.
- W2798928049 hasConcept C33923547 @default.
- W2798928049 hasConcept C41008148 @default.
- W2798928049 hasConcept C49781872 @default.
- W2798928049 hasConcept C49937458 @default.
- W2798928049 hasConcept C61224824 @default.
- W2798928049 hasConcept C75553542 @default.
- W2798928049 hasConcept C9810830 @default.
- W2798928049 hasConceptScore W2798928049C105795698 @default.
- W2798928049 hasConceptScore W2798928049C111219384 @default.
- W2798928049 hasConceptScore W2798928049C111472728 @default.
- W2798928049 hasConceptScore W2798928049C138885662 @default.
- W2798928049 hasConceptScore W2798928049C153180895 @default.
- W2798928049 hasConceptScore W2798928049C154945302 @default.
- W2798928049 hasConceptScore W2798928049C27438332 @default.
- W2798928049 hasConceptScore W2798928049C2775852435 @default.
- W2798928049 hasConceptScore W2798928049C28490314 @default.
- W2798928049 hasConceptScore W2798928049C33923547 @default.
- W2798928049 hasConceptScore W2798928049C41008148 @default.
- W2798928049 hasConceptScore W2798928049C49781872 @default.
- W2798928049 hasConceptScore W2798928049C49937458 @default.
- W2798928049 hasConceptScore W2798928049C61224824 @default.
- W2798928049 hasConceptScore W2798928049C75553542 @default.
- W2798928049 hasConceptScore W2798928049C9810830 @default.
- W2798928049 hasLocation W27989280491 @default.
- W2798928049 hasOpenAccess W2798928049 @default.
- W2798928049 hasPrimaryLocation W27989280491 @default.
- W2798928049 hasRelatedWork W1554840114 @default.
- W2798928049 hasRelatedWork W1839552948 @default.
- W2798928049 hasRelatedWork W1902218045 @default.
- W2798928049 hasRelatedWork W1980229494 @default.
- W2798928049 hasRelatedWork W1999651846 @default.
- W2798928049 hasRelatedWork W2006409748 @default.
- W2798928049 hasRelatedWork W2006566011 @default.
- W2798928049 hasRelatedWork W2011229062 @default.
- W2798928049 hasRelatedWork W2058732030 @default.
- W2798928049 hasRelatedWork W2127672898 @default.
- W2798928049 hasRelatedWork W2187387698 @default.
- W2798928049 hasRelatedWork W2293017759 @default.
- W2798928049 hasRelatedWork W2389673752 @default.
- W2798928049 hasRelatedWork W2393282889 @default.
- W2798928049 hasRelatedWork W2545246288 @default.
- W2798928049 hasRelatedWork W2610211568 @default.
- W2798928049 hasRelatedWork W2943907267 @default.
- W2798928049 hasRelatedWork W2962767738 @default.
- W2798928049 hasRelatedWork W3139985716 @default.
- W2798928049 hasRelatedWork W67063673 @default.
- W2798928049 isParatext "false" @default.
- W2798928049 isRetracted "false" @default.
- W2798928049 magId "2798928049" @default.
- W2798928049 workType "article" @default.