Matches in SemOpenAlex for { <https://semopenalex.org/work/W2399016933> ?p ?o ?g. }
Showing items 1 to 84 of
84
with 100 items per page.
- W2399016933 abstract "In this study, transfer learning techniques are presented for cross-lingual speech recognition to mitigate the effects of limited availability of data in a target language using data from richly resourced source languages. First, a maximum likelihood (ML) based regularization criterion is used to learn context-dependent Gaussian mixture model (GMM) based hidden Markov model (HMM) parameters for phones in target language using data from both target and source languages. Recognition results indicate improved HMM state alignments. Second, the hidden layers of a deep neural network (DNN) are initialized using unsupervised pre-training of a multilingual deep belief network (DBN). The DNN is fine-tuned jointly using a modified cross entropy criterion that uses HMM state alignments from both target and source languages. Third, another DNN fine-tuning technique is explored where the training is performed in a sequential manner source language followed by the target language. Experiments conducted using varying amounts of target data indicate further improvements in performance can be obtained using joint and sequential training of the DNN compared to existing techniques. Turkish and English were chosen to be the target and source languages respectively." @default.
- W2399016933 created "2016-06-24" @default.
- W2399016933 creator A5004778663 @default.
- W2399016933 creator A5060823612 @default.
- W2399016933 date "2015-09-06" @default.
- W2399016933 modified "2023-09-27" @default.
- W2399016933 title "Cross-lingual transfer learning during supervised training in low resource scenarios" @default.
- W2399016933 cites W1524333225 @default.
- W2399016933 cites W1994606281 @default.
- W2399016933 cites W2002342963 @default.
- W2399016933 cites W2012897754 @default.
- W2399016933 cites W2025198378 @default.
- W2399016933 cites W2033436836 @default.
- W2399016933 cites W2105354660 @default.
- W2399016933 cites W2120209245 @default.
- W2399016933 cites W2124629003 @default.
- W2399016933 cites W2125838338 @default.
- W2399016933 cites W2127982613 @default.
- W2399016933 cites W2131042651 @default.
- W2399016933 cites W2141403369 @default.
- W2399016933 cites W2147768505 @default.
- W2399016933 cites W2167771929 @default.
- W2399016933 cites W2307960051 @default.
- W2399016933 cites W2407897255 @default.
- W2399016933 cites W319941341 @default.
- W2399016933 cites W586333776 @default.
- W2399016933 cites W73572011 @default.
- W2399016933 cites W9388558 @default.
- W2399016933 doi "https://doi.org/10.21437/interspeech.2015-700" @default.
- W2399016933 hasPublicationYear "2015" @default.
- W2399016933 type Work @default.
- W2399016933 sameAs 2399016933 @default.
- W2399016933 citedByCount "19" @default.
- W2399016933 countsByYear W23990169332016 @default.
- W2399016933 countsByYear W23990169332017 @default.
- W2399016933 countsByYear W23990169332018 @default.
- W2399016933 countsByYear W23990169332019 @default.
- W2399016933 countsByYear W23990169332020 @default.
- W2399016933 countsByYear W23990169332021 @default.
- W2399016933 countsByYear W23990169332022 @default.
- W2399016933 crossrefType "proceedings-article" @default.
- W2399016933 hasAuthorship W2399016933A5004778663 @default.
- W2399016933 hasAuthorship W2399016933A5060823612 @default.
- W2399016933 hasConcept C119857082 @default.
- W2399016933 hasConcept C137293760 @default.
- W2399016933 hasConcept C150899416 @default.
- W2399016933 hasConcept C153180895 @default.
- W2399016933 hasConcept C154945302 @default.
- W2399016933 hasConcept C204321447 @default.
- W2399016933 hasConcept C23224414 @default.
- W2399016933 hasConcept C2776135515 @default.
- W2399016933 hasConcept C28490314 @default.
- W2399016933 hasConcept C41008148 @default.
- W2399016933 hasConcept C50644808 @default.
- W2399016933 hasConcept C61224824 @default.
- W2399016933 hasConceptScore W2399016933C119857082 @default.
- W2399016933 hasConceptScore W2399016933C137293760 @default.
- W2399016933 hasConceptScore W2399016933C150899416 @default.
- W2399016933 hasConceptScore W2399016933C153180895 @default.
- W2399016933 hasConceptScore W2399016933C154945302 @default.
- W2399016933 hasConceptScore W2399016933C204321447 @default.
- W2399016933 hasConceptScore W2399016933C23224414 @default.
- W2399016933 hasConceptScore W2399016933C2776135515 @default.
- W2399016933 hasConceptScore W2399016933C28490314 @default.
- W2399016933 hasConceptScore W2399016933C41008148 @default.
- W2399016933 hasConceptScore W2399016933C50644808 @default.
- W2399016933 hasConceptScore W2399016933C61224824 @default.
- W2399016933 hasLocation W23990169331 @default.
- W2399016933 hasOpenAccess W2399016933 @default.
- W2399016933 hasPrimaryLocation W23990169331 @default.
- W2399016933 hasRelatedWork W2945992231 @default.
- W2399016933 hasRelatedWork W2946016983 @default.
- W2399016933 hasRelatedWork W2960456850 @default.
- W2399016933 hasRelatedWork W3031818154 @default.
- W2399016933 hasRelatedWork W3131673289 @default.
- W2399016933 hasRelatedWork W3133293092 @default.
- W2399016933 hasRelatedWork W4213299466 @default.
- W2399016933 hasRelatedWork W4225294552 @default.
- W2399016933 hasRelatedWork W4281382123 @default.
- W2399016933 hasRelatedWork W4281645081 @default.
- W2399016933 isParatext "false" @default.
- W2399016933 isRetracted "false" @default.
- W2399016933 magId "2399016933" @default.
- W2399016933 workType "article" @default.