Matches in SemOpenAlex for { <https://semopenalex.org/work/W139772320> ?p ?o ?g. }
Showing items 1 to 70 of
70
with 100 items per page.
- W139772320 abstract "Large vocabulary continuous speech recognition is always a difficult task, and it is particularly so for low-resource languages. The scenario we focus on here is having only 1 hour of acoustic training data in the “target” language. This paper presents work on a data borrowing strategy combined with the recently proposed Subspace Gaussian Mixture Model (SGMM). We developed data borrowing strategies based on two approaches: one based on minimizing K-L Divergence, and one that also takes into account state occupation counts. We demonstrate improvements versus the baseline SGMM setup, which itself is better than a conventional HMM-GMM system. The SGMMs are more robustly estimated by borrowing data from the non-target language at the acousticstate level. Although we tested the approach for SGMMs, we expect the general idea of borrowing data from a non-target language to be applicable for conventional GMMs as well. Index Terms: speech recognition, low-resource language, subspace gaussian mixture model" @default.
- W139772320 created "2016-06-24" @default.
- W139772320 creator A5015824704 @default.
- W139772320 creator A5057117256 @default.
- W139772320 creator A5084286453 @default.
- W139772320 date "2011-08-27" @default.
- W139772320 modified "2023-09-26" @default.
- W139772320 title "State-level data borrowing for low-resource speech recognition based on subspace GMMs" @default.
- W139772320 cites W1631260214 @default.
- W139772320 cites W1932968309 @default.
- W139772320 cites W1981706894 @default.
- W139772320 cites W2000688218 @default.
- W139772320 cites W2123798005 @default.
- W139772320 cites W2130414229 @default.
- W139772320 cites W2151402979 @default.
- W139772320 cites W2152195628 @default.
- W139772320 cites W2164499211 @default.
- W139772320 cites W2165436104 @default.
- W139772320 cites W2166637769 @default.
- W139772320 cites W2170306297 @default.
- W139772320 cites W28619450 @default.
- W139772320 cites W39446388 @default.
- W139772320 doi "https://doi.org/10.21437/interspeech.2011-229" @default.
- W139772320 hasPublicationYear "2011" @default.
- W139772320 type Work @default.
- W139772320 sameAs 139772320 @default.
- W139772320 citedByCount "22" @default.
- W139772320 countsByYear W1397723202012 @default.
- W139772320 countsByYear W1397723202013 @default.
- W139772320 countsByYear W1397723202014 @default.
- W139772320 countsByYear W1397723202015 @default.
- W139772320 countsByYear W1397723202016 @default.
- W139772320 countsByYear W1397723202022 @default.
- W139772320 crossrefType "proceedings-article" @default.
- W139772320 hasAuthorship W139772320A5015824704 @default.
- W139772320 hasAuthorship W139772320A5057117256 @default.
- W139772320 hasAuthorship W139772320A5084286453 @default.
- W139772320 hasConcept C11413529 @default.
- W139772320 hasConcept C153180895 @default.
- W139772320 hasConcept C154945302 @default.
- W139772320 hasConcept C28490314 @default.
- W139772320 hasConcept C32834561 @default.
- W139772320 hasConcept C41008148 @default.
- W139772320 hasConcept C48103436 @default.
- W139772320 hasConcept C61224824 @default.
- W139772320 hasConceptScore W139772320C11413529 @default.
- W139772320 hasConceptScore W139772320C153180895 @default.
- W139772320 hasConceptScore W139772320C154945302 @default.
- W139772320 hasConceptScore W139772320C28490314 @default.
- W139772320 hasConceptScore W139772320C32834561 @default.
- W139772320 hasConceptScore W139772320C41008148 @default.
- W139772320 hasConceptScore W139772320C48103436 @default.
- W139772320 hasConceptScore W139772320C61224824 @default.
- W139772320 hasLocation W1397723201 @default.
- W139772320 hasOpenAccess W139772320 @default.
- W139772320 hasPrimaryLocation W1397723201 @default.
- W139772320 hasRelatedWork W1998938004 @default.
- W139772320 hasRelatedWork W2016349419 @default.
- W139772320 hasRelatedWork W2048392085 @default.
- W139772320 hasRelatedWork W2119875513 @default.
- W139772320 hasRelatedWork W2157903613 @default.
- W139772320 hasRelatedWork W2159633528 @default.
- W139772320 hasRelatedWork W2321141263 @default.
- W139772320 hasRelatedWork W2543161807 @default.
- W139772320 hasRelatedWork W2164844972 @default.
- W139772320 hasRelatedWork W2171745948 @default.
- W139772320 isParatext "false" @default.
- W139772320 isRetracted "false" @default.
- W139772320 magId "139772320" @default.
- W139772320 workType "article" @default.