FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2095635231> ?p ?o ?g. }

• W2095635231 endingPage "241" @default.
• W2095635231 startingPage "225" @default.
• W2095635231 abstract "We analyze a simple hierarchical architecture consisting of two multilayer perceptron (MLP) classifiers in tandem to estimate the phonetic class conditional probabilities. In this hierarchical setup, the first MLP classifier is trained using standard acoustic features. The second MLP is trained using the posterior probabilities of phonemes estimated by the first, but with a long temporal context of around 150-230 ms. Through extensive phoneme recognition experiments, and the analysis of the trained second MLP using Volterra series, we show that 1) the hierarchical system yields higher phoneme recognition accuracies-an absolute improvement of 3.5% and 9.3% on TIMIT and CTS respectively-over the conventional single MLP-based system, 2) there exists useful information in the temporal trajectories of the posterior feature space, spanning around 230 ms of context, 3) the second MLP learns the phonetic temporal patterns in the posterior features, which include the phonetic confusions at the output of the first MLP as well as the phonotactics of the language as observed in the training data, and 4) the second MLP classifier requires fewer number of parameters and can be trained using lesser amount of training data." @default.
• W2095635231 created "2016-06-24" @default.
• W2095635231 creator A5042260050 @default.
• W2095635231 creator A5047712456 @default.
• W2095635231 creator A5064656096 @default.
• W2095635231 creator A5065376601 @default.
• W2095635231 creator A5067669848 @default.
• W2095635231 date "2011-02-01" @default.
• W2095635231 modified "2023-10-17" @default.
• W2095635231 title "Analysis of MLP-Based Hierarchical Phoneme Posterior Probability Estimator" @default.
• W2095635231 cites W132821814 @default.
• W2095635231 cites W1504129048 @default.
• W2095635231 cites W1534536260 @default.
• W2095635231 cites W197151588 @default.
• W2095635231 cites W1977152366 @default.
• W2095635231 cites W1980501707 @default.
• W2095635231 cites W2071489795 @default.
• W2095635231 cites W2077804127 @default.
• W2095635231 cites W2095635999 @default.
• W2095635231 cites W2097560587 @default.
• W2095635231 cites W2100932383 @default.
• W2095635231 cites W2101596234 @default.
• W2095635231 cites W2102512139 @default.
• W2095635231 cites W2103372210 @default.
• W2095635231 cites W2105897600 @default.
• W2095635231 cites W2109426048 @default.
• W2095635231 cites W2110871230 @default.
• W2095635231 cites W2114665324 @default.
• W2095635231 cites W2115407392 @default.
• W2095635231 cites W2118858754 @default.
• W2095635231 cites W2125534887 @default.
• W2095635231 cites W2125610823 @default.
• W2095635231 cites W2128383815 @default.
• W2095635231 cites W2133924365 @default.
• W2095635231 cites W2134900903 @default.
• W2095635231 cites W2141499240 @default.
• W2095635231 cites W2143137842 @default.
• W2095635231 cites W2153839310 @default.
• W2095635231 cites W2161481180 @default.
• W2095635231 cites W2166637769 @default.
• W2095635231 cites W2167535584 @default.
• W2095635231 cites W2167763959 @default.
• W2095635231 cites W2499601348 @default.
• W2095635231 cites W2546285092 @default.
• W2095635231 cites W3145614364 @default.
• W2095635231 cites W3149335959 @default.
• W2095635231 cites W4238404964 @default.
• W2095635231 cites W811578723 @default.
• W2095635231 doi "https://doi.org/10.1109/tasl.2010.2045943" @default.
• W2095635231 hasPublicationYear "2011" @default.
• W2095635231 type Work @default.
• W2095635231 sameAs 2095635231 @default.
• W2095635231 citedByCount "74" @default.
• W2095635231 countsByYear W20956352312012 @default.
• W2095635231 countsByYear W20956352312013 @default.
• W2095635231 countsByYear W20956352312014 @default.
• W2095635231 countsByYear W20956352312015 @default.
• W2095635231 countsByYear W20956352312016 @default.
• W2095635231 countsByYear W20956352312017 @default.
• W2095635231 countsByYear W20956352312018 @default.
• W2095635231 countsByYear W20956352312019 @default.
• W2095635231 countsByYear W20956352312020 @default.
• W2095635231 countsByYear W20956352312021 @default.
• W2095635231 countsByYear W20956352312022 @default.
• W2095635231 countsByYear W20956352312023 @default.
• W2095635231 crossrefType "journal-article" @default.
• W2095635231 hasAuthorship W2095635231A5042260050 @default.
• W2095635231 hasAuthorship W2095635231A5047712456 @default.
• W2095635231 hasAuthorship W2095635231A5064656096 @default.
• W2095635231 hasAuthorship W2095635231A5065376601 @default.
• W2095635231 hasAuthorship W2095635231A5067669848 @default.
• W2095635231 hasBestOaLocation W20956352312 @default.
• W2095635231 hasConcept C107673813 @default.
• W2095635231 hasConcept C138885662 @default.
• W2095635231 hasConcept C148934300 @default.
• W2095635231 hasConcept C153180895 @default.
• W2095635231 hasConcept C154945302 @default.
• W2095635231 hasConcept C179717631 @default.
• W2095635231 hasConcept C23224414 @default.
• W2095635231 hasConcept C2778724510 @default.
• W2095635231 hasConcept C28490314 @default.
• W2095635231 hasConcept C41008148 @default.
• W2095635231 hasConcept C41895202 @default.
• W2095635231 hasConcept C4768521 @default.
• W2095635231 hasConcept C50644808 @default.
• W2095635231 hasConcept C57830394 @default.
• W2095635231 hasConcept C83665646 @default.
• W2095635231 hasConcept C95623464 @default.
• W2095635231 hasConceptScore W2095635231C107673813 @default.
• W2095635231 hasConceptScore W2095635231C138885662 @default.
• W2095635231 hasConceptScore W2095635231C148934300 @default.
• W2095635231 hasConceptScore W2095635231C153180895 @default.
• W2095635231 hasConceptScore W2095635231C154945302 @default.
• W2095635231 hasConceptScore W2095635231C179717631 @default.
• W2095635231 hasConceptScore W2095635231C23224414 @default.
• W2095635231 hasConceptScore W2095635231C2778724510 @default.
• W2095635231 hasConceptScore W2095635231C28490314 @default.
• W2095635231 hasConceptScore W2095635231C41008148 @default.

• next