Matches in SemOpenAlex for { <https://semopenalex.org/work/W2140372979> ?p ?o ?g. }
- W2140372979 abstract "Generation of high-precision sub-phonetic attribute (also known as phonological features) and phone lattices is a key frontend component for detection-based bottom-up speech recognition. In this paper we employ deep neural networks (DNNs) to improve detection accuracy over conventional shallow MLPs (multi-layer perceptrons) with one hidden layer. A range of DNN architectures with five to seven hidden layers and up to 2048 hidden units per layer have been explored. Training on the SI84 and testing on the Nov92 WSJ data, the proposed DNNs achieve significant improvements over the shallow MLPs, producing greater than 90% frame-level attribute estimation accuracies for all 21 attributes tested for the full system. On the phone detection task, we also obtain excellent frame-level accuracy of 86.6%. With this level of high-precision detection of basic speech units we have opened the door to a new family of flexible speech recognition system design for both top-down and bottom-up, lattice-based search strategies and knowledge integration." @default.
- W2140372979 created "2016-06-24" @default.
- W2140372979 creator A5034476404 @default.
- W2140372979 creator A5035017068 @default.
- W2140372979 creator A5066868860 @default.
- W2140372979 creator A5079659476 @default.
- W2140372979 date "2012-03-01" @default.
- W2140372979 modified "2023-09-24" @default.
- W2140372979 title "Boosting attribute and phone estimation accuracies with deep neural networks for detection-based speech recognition" @default.
- W2140372979 cites W1490506669 @default.
- W2140372979 cites W1583757088 @default.
- W2140372979 cites W173660884 @default.
- W2140372979 cites W1784780341 @default.
- W2140372979 cites W1975598412 @default.
- W2140372979 cites W1980850109 @default.
- W2140372979 cites W1993882792 @default.
- W2140372979 cites W2024490156 @default.
- W2140372979 cites W2053280194 @default.
- W2140372979 cites W2060630096 @default.
- W2140372979 cites W2116064496 @default.
- W2140372979 cites W2123237149 @default.
- W2140372979 cites W2141133575 @default.
- W2140372979 cites W2141354973 @default.
- W2140372979 cites W2141778357 @default.
- W2140372979 cites W2147768505 @default.
- W2140372979 cites W2151311453 @default.
- W2140372979 cites W2152195628 @default.
- W2140372979 cites W2160373860 @default.
- W2140372979 cites W2171670492 @default.
- W2140372979 cites W217970951 @default.
- W2140372979 cites W2394932179 @default.
- W2140372979 cites W2405868127 @default.
- W2140372979 cites W271462270 @default.
- W2140372979 cites W29056978 @default.
- W2140372979 cites W2074642696 @default.
- W2140372979 doi "https://doi.org/10.1109/icassp.2012.6288837" @default.
- W2140372979 hasPublicationYear "2012" @default.
- W2140372979 type Work @default.
- W2140372979 sameAs 2140372979 @default.
- W2140372979 citedByCount "56" @default.
- W2140372979 countsByYear W21403729792012 @default.
- W2140372979 countsByYear W21403729792013 @default.
- W2140372979 countsByYear W21403729792014 @default.
- W2140372979 countsByYear W21403729792015 @default.
- W2140372979 countsByYear W21403729792016 @default.
- W2140372979 countsByYear W21403729792017 @default.
- W2140372979 countsByYear W21403729792018 @default.
- W2140372979 countsByYear W21403729792019 @default.
- W2140372979 countsByYear W21403729792020 @default.
- W2140372979 countsByYear W21403729792021 @default.
- W2140372979 crossrefType "proceedings-article" @default.
- W2140372979 hasAuthorship W2140372979A5034476404 @default.
- W2140372979 hasAuthorship W2140372979A5035017068 @default.
- W2140372979 hasAuthorship W2140372979A5066868860 @default.
- W2140372979 hasAuthorship W2140372979A5079659476 @default.
- W2140372979 hasConcept C126042441 @default.
- W2140372979 hasConcept C138885662 @default.
- W2140372979 hasConcept C153180895 @default.
- W2140372979 hasConcept C154945302 @default.
- W2140372979 hasConcept C179717631 @default.
- W2140372979 hasConcept C204201278 @default.
- W2140372979 hasConcept C23224414 @default.
- W2140372979 hasConcept C2778707766 @default.
- W2140372979 hasConcept C28490314 @default.
- W2140372979 hasConcept C41008148 @default.
- W2140372979 hasConcept C41895202 @default.
- W2140372979 hasConcept C46686674 @default.
- W2140372979 hasConcept C50644808 @default.
- W2140372979 hasConcept C60908668 @default.
- W2140372979 hasConcept C61328038 @default.
- W2140372979 hasConcept C76155785 @default.
- W2140372979 hasConceptScore W2140372979C126042441 @default.
- W2140372979 hasConceptScore W2140372979C138885662 @default.
- W2140372979 hasConceptScore W2140372979C153180895 @default.
- W2140372979 hasConceptScore W2140372979C154945302 @default.
- W2140372979 hasConceptScore W2140372979C179717631 @default.
- W2140372979 hasConceptScore W2140372979C204201278 @default.
- W2140372979 hasConceptScore W2140372979C23224414 @default.
- W2140372979 hasConceptScore W2140372979C2778707766 @default.
- W2140372979 hasConceptScore W2140372979C28490314 @default.
- W2140372979 hasConceptScore W2140372979C41008148 @default.
- W2140372979 hasConceptScore W2140372979C41895202 @default.
- W2140372979 hasConceptScore W2140372979C46686674 @default.
- W2140372979 hasConceptScore W2140372979C50644808 @default.
- W2140372979 hasConceptScore W2140372979C60908668 @default.
- W2140372979 hasConceptScore W2140372979C61328038 @default.
- W2140372979 hasConceptScore W2140372979C76155785 @default.
- W2140372979 hasLocation W21403729791 @default.
- W2140372979 hasOpenAccess W2140372979 @default.
- W2140372979 hasPrimaryLocation W21403729791 @default.
- W2140372979 hasRelatedWork W1920431333 @default.
- W2140372979 hasRelatedWork W1967332585 @default.
- W2140372979 hasRelatedWork W2019307865 @default.
- W2140372979 hasRelatedWork W2105626690 @default.
- W2140372979 hasRelatedWork W2140372979 @default.
- W2140372979 hasRelatedWork W2182829855 @default.
- W2140372979 hasRelatedWork W2185910246 @default.
- W2140372979 hasRelatedWork W2314526909 @default.
- W2140372979 hasRelatedWork W294719094 @default.
- W2140372979 hasRelatedWork W2109868656 @default.