FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1513984940> ?p ?o ?g. }

• W1513984940 abstract "A large body of research has shown that acoustic features for speech recognition can be learned from data using neural networks with multiple hidden layers (DNNs) and that these learned features are superior to standard features (e.g., MFCCs). However, this superiority is usually demonstrated when the data used to learn the features is very similar in character to the data used to test recognition performance. An open question is how well these learned features generalize to realistic data that is different in character to their training data. The ability of a feature representation to generalize to unfamiliar data is a highly desirable form of robustness. In this paper we investigate the robustness of two DNN-based feature sets to training/test mismatch using the ICSI meeting corpus. The experiments were performed under 3 training/test scenarios: (1) matched near-field (2) matched far-field and (3) the mismatched condition near-field training with far-field testing. The experiments leverage simulation and a novel sampling process that we have developed for diagnostic analysis within the HMM-based speech recognition framework. First, diagnostic analysis shows that a DNN-based feature representation that uses MFCC inputs (MFCC-DNN) is indeed superior to the corresponding MFCC baselines in the two matched scenarios where the source of recognition errors are from incorrect model, but the DNN-based features and MFCCs have nearly identical and poor performance in the mismatched scenario. Second, we show that a DNN-based feature representation that uses a more robust input, namely power normalized spectrum (PNS) and Gabor filters, performs nearly as well as the MFCC-DNN features in the matched scenarios and much better than MFCCs and MFCC-DNNs in the mismatched scenario." @default.
• W1513984940 created "2016-06-24" @default.
• W1513984940 creator A5001306222 @default.
• W1513984940 creator A5025957537 @default.
• W1513984940 date "2015-04-01" @default.
• W1513984940 modified "2023-09-26" @default.
• W1513984940 title "On the importance of modeling and robustness for deep neural network feature" @default.
• W1513984940 cites W1553004968 @default.
• W1513984940 cites W1591607137 @default.
• W1513984940 cites W1627087495 @default.
• W1513984940 cites W197151588 @default.
• W1513984940 cites W1978741356 @default.
• W1513984940 cites W1993882792 @default.
• W1513984940 cites W2030937226 @default.
• W1513984940 cites W2048608953 @default.
• W1513984940 cites W2072128103 @default.
• W1513984940 cites W2072349636 @default.
• W1513984940 cites W2088489891 @default.
• W1513984940 cites W2113427327 @default.
• W1513984940 cites W2121573996 @default.
• W1513984940 cites W2143612262 @default.
• W1513984940 cites W2147768505 @default.
• W1513984940 cites W2160815625 @default.
• W1513984940 cites W2165712214 @default.
• W1513984940 cites W2188183693 @default.
• W1513984940 cites W2293634267 @default.
• W1513984940 cites W2395106899 @default.
• W1513984940 cites W75470185 @default.
• W1513984940 doi "https://doi.org/10.1109/icassp.2015.7178828" @default.
• W1513984940 hasPublicationYear "2015" @default.
• W1513984940 type Work @default.
• W1513984940 sameAs 1513984940 @default.
• W1513984940 citedByCount "10" @default.
• W1513984940 countsByYear W15139849402015 @default.
• W1513984940 countsByYear W15139849402016 @default.
• W1513984940 countsByYear W15139849402017 @default.
• W1513984940 countsByYear W15139849402018 @default.
• W1513984940 countsByYear W15139849402019 @default.
• W1513984940 crossrefType "proceedings-article" @default.
• W1513984940 hasAuthorship W1513984940A5001306222 @default.
• W1513984940 hasAuthorship W1513984940A5025957537 @default.
• W1513984940 hasConcept C104317684 @default.
• W1513984940 hasConcept C138885662 @default.
• W1513984940 hasConcept C151989614 @default.
• W1513984940 hasConcept C153083717 @default.
• W1513984940 hasConcept C153180895 @default.
• W1513984940 hasConcept C154945302 @default.
• W1513984940 hasConcept C16910744 @default.
• W1513984940 hasConcept C185592680 @default.
• W1513984940 hasConcept C199360897 @default.
• W1513984940 hasConcept C23224414 @default.
• W1513984940 hasConcept C2776401178 @default.
• W1513984940 hasConcept C28490314 @default.
• W1513984940 hasConcept C41008148 @default.
• W1513984940 hasConcept C41895202 @default.
• W1513984940 hasConcept C50644808 @default.
• W1513984940 hasConcept C52622490 @default.
• W1513984940 hasConcept C55493867 @default.
• W1513984940 hasConcept C63479239 @default.
• W1513984940 hasConceptScore W1513984940C104317684 @default.
• W1513984940 hasConceptScore W1513984940C138885662 @default.
• W1513984940 hasConceptScore W1513984940C151989614 @default.
• W1513984940 hasConceptScore W1513984940C153083717 @default.
• W1513984940 hasConceptScore W1513984940C153180895 @default.
• W1513984940 hasConceptScore W1513984940C154945302 @default.
• W1513984940 hasConceptScore W1513984940C16910744 @default.
• W1513984940 hasConceptScore W1513984940C185592680 @default.
• W1513984940 hasConceptScore W1513984940C199360897 @default.
• W1513984940 hasConceptScore W1513984940C23224414 @default.
• W1513984940 hasConceptScore W1513984940C2776401178 @default.
• W1513984940 hasConceptScore W1513984940C28490314 @default.
• W1513984940 hasConceptScore W1513984940C41008148 @default.
• W1513984940 hasConceptScore W1513984940C41895202 @default.
• W1513984940 hasConceptScore W1513984940C50644808 @default.
• W1513984940 hasConceptScore W1513984940C52622490 @default.
• W1513984940 hasConceptScore W1513984940C55493867 @default.
• W1513984940 hasConceptScore W1513984940C63479239 @default.
• W1513984940 hasLocation W15139849401 @default.
• W1513984940 hasOpenAccess W1513984940 @default.
• W1513984940 hasPrimaryLocation W15139849401 @default.
• W1513984940 hasRelatedWork W107532267 @default.
• W1513984940 hasRelatedWork W1976677479 @default.
• W1513984940 hasRelatedWork W2015458579 @default.
• W1513984940 hasRelatedWork W2037552098 @default.
• W1513984940 hasRelatedWork W2080758042 @default.
• W1513984940 hasRelatedWork W2127377615 @default.
• W1513984940 hasRelatedWork W2147768505 @default.
• W1513984940 hasRelatedWork W2161224286 @default.
• W1513984940 hasRelatedWork W2161678726 @default.
• W1513984940 hasRelatedWork W2171670492 @default.
• W1513984940 hasRelatedWork W2345067732 @default.
• W1513984940 hasRelatedWork W2406262283 @default.
• W1513984940 hasRelatedWork W2547335510 @default.
• W1513984940 hasRelatedWork W2578331995 @default.
• W1513984940 hasRelatedWork W2604352297 @default.
• W1513984940 hasRelatedWork W2890351244 @default.
• W1513984940 hasRelatedWork W2895807593 @default.
• W1513984940 hasRelatedWork W2980919576 @default.
• W1513984940 hasRelatedWork W3135915893 @default.
• W1513984940 hasRelatedWork W3137430505 @default.

• next