Matches in SemOpenAlex for { <https://semopenalex.org/work/W1547774380> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W1547774380 abstract "Discriminative classifiers are a popular approach to solving classification problems. However one of the problems with these approaches, in particular kernel based classifiers such as Support Vector Machines (SVMs), is that they are hard to adapt to mismatches between the training and test data. This paper describes a scheme for overcoming this problem for speech recognition in noise. Generative kernels, defined using generative models, allow SVMs to handle sequence data. By compensating the generative models for the noise conditions noise-specific generative kernels can be obtained. These can be used to train a noise-independent SVM on a range of noise conditions, which can then be used with a test-set noise kernel for classification. Initial experiments using an idealised version of model-based compensation were run on the AURORA 2.0 continuous digit task. The proposed scheme yielded large gains in performance over the compensated models." @default.
- W1547774380 created "2016-06-24" @default.
- W1547774380 creator A5012703069 @default.
- W1547774380 creator A5019581856 @default.
- W1547774380 date "2008-09-22" @default.
- W1547774380 modified "2023-09-26" @default.
- W1547774380 title "Discriminative classifiers with generative kernels for noise robust ASR" @default.
- W1547774380 cites W105750483 @default.
- W1547774380 cites W1487784522 @default.
- W1547774380 cites W1665196592 @default.
- W1547774380 cites W2067117291 @default.
- W1547774380 cites W2080921589 @default.
- W1547774380 cites W2137401359 @default.
- W1547774380 cites W2140409019 @default.
- W1547774380 cites W2148603752 @default.
- W1547774380 cites W2161213733 @default.
- W1547774380 cites W2162042984 @default.
- W1547774380 cites W2280467666 @default.
- W1547774380 doi "https://doi.org/10.21437/interspeech.2008-317" @default.
- W1547774380 hasPublicationYear "2008" @default.
- W1547774380 type Work @default.
- W1547774380 sameAs 1547774380 @default.
- W1547774380 citedByCount "6" @default.
- W1547774380 crossrefType "proceedings-article" @default.
- W1547774380 hasAuthorship W1547774380A5012703069 @default.
- W1547774380 hasAuthorship W1547774380A5019581856 @default.
- W1547774380 hasBestOaLocation W15477743802 @default.
- W1547774380 hasConcept C115961682 @default.
- W1547774380 hasConcept C119857082 @default.
- W1547774380 hasConcept C153180895 @default.
- W1547774380 hasConcept C154945302 @default.
- W1547774380 hasConcept C167966045 @default.
- W1547774380 hasConcept C28490314 @default.
- W1547774380 hasConcept C39890363 @default.
- W1547774380 hasConcept C41008148 @default.
- W1547774380 hasConcept C97931131 @default.
- W1547774380 hasConcept C99498987 @default.
- W1547774380 hasConceptScore W1547774380C115961682 @default.
- W1547774380 hasConceptScore W1547774380C119857082 @default.
- W1547774380 hasConceptScore W1547774380C153180895 @default.
- W1547774380 hasConceptScore W1547774380C154945302 @default.
- W1547774380 hasConceptScore W1547774380C167966045 @default.
- W1547774380 hasConceptScore W1547774380C28490314 @default.
- W1547774380 hasConceptScore W1547774380C39890363 @default.
- W1547774380 hasConceptScore W1547774380C41008148 @default.
- W1547774380 hasConceptScore W1547774380C97931131 @default.
- W1547774380 hasConceptScore W1547774380C99498987 @default.
- W1547774380 hasLocation W15477743801 @default.
- W1547774380 hasLocation W15477743802 @default.
- W1547774380 hasOpenAccess W1547774380 @default.
- W1547774380 hasPrimaryLocation W15477743801 @default.
- W1547774380 hasRelatedWork W1534961803 @default.
- W1547774380 hasRelatedWork W2096034952 @default.
- W1547774380 hasRelatedWork W2137117485 @default.
- W1547774380 hasRelatedWork W2592408322 @default.
- W1547774380 hasRelatedWork W2770426046 @default.
- W1547774380 hasRelatedWork W2991061294 @default.
- W1547774380 hasRelatedWork W2994891734 @default.
- W1547774380 hasRelatedWork W4302800336 @default.
- W1547774380 hasRelatedWork W4310424780 @default.
- W1547774380 hasRelatedWork W2310403681 @default.
- W1547774380 isParatext "false" @default.
- W1547774380 isRetracted "false" @default.
- W1547774380 magId "1547774380" @default.
- W1547774380 workType "article" @default.