Matches in SemOpenAlex for { <https://semopenalex.org/work/W873631249> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W873631249 endingPage "80" @default.
- W873631249 startingPage "71" @default.
- W873631249 abstract "Robust acoustic modeling is essential in the development of automatic speech recognition systems applied to spoken human-computer interaction. To this end, traditional hidden Markov models (HMM) may be improved by hybridizing them with artificial neural networks (ANN). Crucially, ANNs require input values that do not compromize their numerical stability. In spite of the relevance feature normalization has on the success of ANNs in real-world applications, the issue is mostly overlooked on the false premize that “any normalization technique will do”. The paper proposes a gradient-ascent, maximum-likelihood algorithm for feature normalization. Relying on mixtures of logistic densities, it ensures ANN-friendly values that are distributed over the (0, 1) interval in a uniform manner. Some nice properties of the approach are discussed. The algorithm is applied to the normalization of acoustic features for a hybrid ANN/HMM speech recognizer. Experiments on real-world continuous speech recognition tasks are presented. The hybrid system turns out to be positively affected by the proposed technique." @default.
- W873631249 created "2016-06-24" @default.
- W873631249 creator A5053948915 @default.
- W873631249 date "2015-11-01" @default.
- W873631249 modified "2023-09-25" @default.
- W873631249 title "Maximum-likelihood normalization of features increases the robustness of neural-based spoken human-computer interaction" @default.
- W873631249 cites W2005679925 @default.
- W873631249 cites W2006638151 @default.
- W873631249 cites W2007857129 @default.
- W873631249 cites W2017743541 @default.
- W873631249 cites W2040118574 @default.
- W873631249 cites W2084585855 @default.
- W873631249 cites W2103496339 @default.
- W873631249 cites W2125838338 @default.
- W873631249 cites W2148154194 @default.
- W873631249 cites W2166993553 @default.
- W873631249 cites W4231956489 @default.
- W873631249 doi "https://doi.org/10.1016/j.patrec.2015.07.003" @default.
- W873631249 hasPublicationYear "2015" @default.
- W873631249 type Work @default.
- W873631249 sameAs 873631249 @default.
- W873631249 citedByCount "7" @default.
- W873631249 countsByYear W8736312492017 @default.
- W873631249 countsByYear W8736312492018 @default.
- W873631249 countsByYear W8736312492020 @default.
- W873631249 countsByYear W8736312492021 @default.
- W873631249 countsByYear W8736312492023 @default.
- W873631249 crossrefType "journal-article" @default.
- W873631249 hasAuthorship W873631249A5053948915 @default.
- W873631249 hasConcept C104317684 @default.
- W873631249 hasConcept C136886441 @default.
- W873631249 hasConcept C138885662 @default.
- W873631249 hasConcept C144024400 @default.
- W873631249 hasConcept C153180895 @default.
- W873631249 hasConcept C154945302 @default.
- W873631249 hasConcept C185592680 @default.
- W873631249 hasConcept C19165224 @default.
- W873631249 hasConcept C23224414 @default.
- W873631249 hasConcept C2776401178 @default.
- W873631249 hasConcept C28490314 @default.
- W873631249 hasConcept C2984842247 @default.
- W873631249 hasConcept C41008148 @default.
- W873631249 hasConcept C41895202 @default.
- W873631249 hasConcept C50644808 @default.
- W873631249 hasConcept C55493867 @default.
- W873631249 hasConcept C63479239 @default.
- W873631249 hasConceptScore W873631249C104317684 @default.
- W873631249 hasConceptScore W873631249C136886441 @default.
- W873631249 hasConceptScore W873631249C138885662 @default.
- W873631249 hasConceptScore W873631249C144024400 @default.
- W873631249 hasConceptScore W873631249C153180895 @default.
- W873631249 hasConceptScore W873631249C154945302 @default.
- W873631249 hasConceptScore W873631249C185592680 @default.
- W873631249 hasConceptScore W873631249C19165224 @default.
- W873631249 hasConceptScore W873631249C23224414 @default.
- W873631249 hasConceptScore W873631249C2776401178 @default.
- W873631249 hasConceptScore W873631249C28490314 @default.
- W873631249 hasConceptScore W873631249C2984842247 @default.
- W873631249 hasConceptScore W873631249C41008148 @default.
- W873631249 hasConceptScore W873631249C41895202 @default.
- W873631249 hasConceptScore W873631249C50644808 @default.
- W873631249 hasConceptScore W873631249C55493867 @default.
- W873631249 hasConceptScore W873631249C63479239 @default.
- W873631249 hasLocation W8736312491 @default.
- W873631249 hasOpenAccess W873631249 @default.
- W873631249 hasPrimaryLocation W8736312491 @default.
- W873631249 hasRelatedWork W105527591 @default.
- W873631249 hasRelatedWork W1650162272 @default.
- W873631249 hasRelatedWork W1851196109 @default.
- W873631249 hasRelatedWork W1986765836 @default.
- W873631249 hasRelatedWork W1991269640 @default.
- W873631249 hasRelatedWork W2011731246 @default.
- W873631249 hasRelatedWork W2016839265 @default.
- W873631249 hasRelatedWork W2097654837 @default.
- W873631249 hasRelatedWork W2510388867 @default.
- W873631249 hasRelatedWork W99997047 @default.
- W873631249 hasVolume "66" @default.
- W873631249 isParatext "false" @default.
- W873631249 isRetracted "false" @default.
- W873631249 magId "873631249" @default.
- W873631249 workType "article" @default.