Matches in SemOpenAlex for { <https://semopenalex.org/work/W111968546> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W111968546 abstract "Recently, there have been intensive studies of margin-based learning for automatic speech recognition (ASR). It is our believe that by securing a margin from the decision boundaries to the training samples, a correct decision can still be made if the mismatches between testing and training samples are well within the tolerance region specified by the margin. This nice property should be effective for robust ASR, where the testing condition is different from those in training. In this paper, we report on experiment results with soft margin estimation (SME) on the Aurora2 task and show that SME is very effective under clean training with more than 50% relative word error reductions in the clean, 20db, and 15db testing conditions, and still gives a slight improvement over conventional multi-condition training approaches. This demonstrates that the margin in SME can equip recognizers with a nice generalization property under adverse conditions." @default.
- W111968546 created "2016-06-24" @default.
- W111968546 creator A5029670581 @default.
- W111968546 creator A5066868860 @default.
- W111968546 date "2008-09-22" @default.
- W111968546 modified "2023-09-26" @default.
- W111968546 title "On a generalization of margin-based discriminative training to robust speech recognition" @default.
- W111968546 cites W1574530145 @default.
- W111968546 cites W2050693797 @default.
- W111968546 cites W2080921589 @default.
- W111968546 cites W2082474452 @default.
- W111968546 cites W2105613180 @default.
- W111968546 cites W2110467295 @default.
- W111968546 cites W2111479622 @default.
- W111968546 cites W2128653836 @default.
- W111968546 cites W2146871184 @default.
- W111968546 cites W2150142469 @default.
- W111968546 cites W2151484683 @default.
- W111968546 cites W2156909104 @default.
- W111968546 cites W2158289097 @default.
- W111968546 cites W2159112514 @default.
- W111968546 cites W84880661 @default.
- W111968546 doi "https://doi.org/10.21437/interspeech.2008-316" @default.
- W111968546 hasPublicationYear "2008" @default.
- W111968546 type Work @default.
- W111968546 sameAs 111968546 @default.
- W111968546 citedByCount "6" @default.
- W111968546 countsByYear W1119685462013 @default.
- W111968546 countsByYear W1119685462019 @default.
- W111968546 crossrefType "proceedings-article" @default.
- W111968546 hasAuthorship W111968546A5029670581 @default.
- W111968546 hasAuthorship W111968546A5066868860 @default.
- W111968546 hasConcept C111472728 @default.
- W111968546 hasConcept C119857082 @default.
- W111968546 hasConcept C121332964 @default.
- W111968546 hasConcept C127413603 @default.
- W111968546 hasConcept C134306372 @default.
- W111968546 hasConcept C138885662 @default.
- W111968546 hasConcept C153180895 @default.
- W111968546 hasConcept C153294291 @default.
- W111968546 hasConcept C154945302 @default.
- W111968546 hasConcept C177148314 @default.
- W111968546 hasConcept C189950617 @default.
- W111968546 hasConcept C201995342 @default.
- W111968546 hasConcept C2777211547 @default.
- W111968546 hasConcept C2780451532 @default.
- W111968546 hasConcept C28490314 @default.
- W111968546 hasConcept C33923547 @default.
- W111968546 hasConcept C41008148 @default.
- W111968546 hasConcept C51632099 @default.
- W111968546 hasConcept C774472 @default.
- W111968546 hasConcept C97931131 @default.
- W111968546 hasConceptScore W111968546C111472728 @default.
- W111968546 hasConceptScore W111968546C119857082 @default.
- W111968546 hasConceptScore W111968546C121332964 @default.
- W111968546 hasConceptScore W111968546C127413603 @default.
- W111968546 hasConceptScore W111968546C134306372 @default.
- W111968546 hasConceptScore W111968546C138885662 @default.
- W111968546 hasConceptScore W111968546C153180895 @default.
- W111968546 hasConceptScore W111968546C153294291 @default.
- W111968546 hasConceptScore W111968546C154945302 @default.
- W111968546 hasConceptScore W111968546C177148314 @default.
- W111968546 hasConceptScore W111968546C189950617 @default.
- W111968546 hasConceptScore W111968546C201995342 @default.
- W111968546 hasConceptScore W111968546C2777211547 @default.
- W111968546 hasConceptScore W111968546C2780451532 @default.
- W111968546 hasConceptScore W111968546C28490314 @default.
- W111968546 hasConceptScore W111968546C33923547 @default.
- W111968546 hasConceptScore W111968546C41008148 @default.
- W111968546 hasConceptScore W111968546C51632099 @default.
- W111968546 hasConceptScore W111968546C774472 @default.
- W111968546 hasConceptScore W111968546C97931131 @default.
- W111968546 hasLocation W1119685461 @default.
- W111968546 hasOpenAccess W111968546 @default.
- W111968546 hasPrimaryLocation W1119685461 @default.
- W111968546 hasRelatedWork W1518381585 @default.
- W111968546 hasRelatedWork W1565453839 @default.
- W111968546 hasRelatedWork W1979727248 @default.
- W111968546 hasRelatedWork W1999415379 @default.
- W111968546 hasRelatedWork W2001679125 @default.
- W111968546 hasRelatedWork W2096734621 @default.
- W111968546 hasRelatedWork W2100969003 @default.
- W111968546 hasRelatedWork W2104591305 @default.
- W111968546 hasRelatedWork W2145192183 @default.
- W111968546 hasRelatedWork W2145815249 @default.
- W111968546 hasRelatedWork W2146871184 @default.
- W111968546 hasRelatedWork W2147606510 @default.
- W111968546 hasRelatedWork W2151484683 @default.
- W111968546 hasRelatedWork W2157421955 @default.
- W111968546 hasRelatedWork W2158289097 @default.
- W111968546 hasRelatedWork W2159112514 @default.
- W111968546 hasRelatedWork W2184410112 @default.
- W111968546 hasRelatedWork W2348395790 @default.
- W111968546 hasRelatedWork W30564313 @default.
- W111968546 hasRelatedWork W5289323 @default.
- W111968546 isParatext "false" @default.
- W111968546 isRetracted "false" @default.
- W111968546 magId "111968546" @default.
- W111968546 workType "article" @default.