Matches in SemOpenAlex for { <https://semopenalex.org/work/W2116724101> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W2116724101 abstract "Latent Dirichlet Allocation (LDA) has been shown to be an effective model to augment n-gram language models in speech recognition applications. In this work, we aim to take advantage of the superior unsupervised learning ability of the framework, and use it to uncover topic structure embedded in the corpora in an entirely data-driven fashion. In addition, we describe a bi-level inference and classification method that allows topic clustering at the utterance level while preserving the document-level topic structures. We demonstrate the effectiveness of the proposed topic clustering pipeline in a state-of-the-art Arabic broadcast transcription system. Experiments show that optimizing LM in the LDA topic space leads to 5% reduction in language model perplexity. It is further shown that topic clustering and adaptation is able to attain 0.4% absolute word error rate reduction on the GALE Arabic task." @default.
- W2116724101 created "2016-06-24" @default.
- W2116724101 creator A5007695606 @default.
- W2116724101 creator A5019973376 @default.
- W2116724101 date "2012-03-01" @default.
- W2116724101 modified "2023-09-27" @default.
- W2116724101 title "Improving arabic broadcast transcription using automatic topic clustering" @default.
- W2116724101 cites W158414620 @default.
- W2116724101 cites W1612003148 @default.
- W2116724101 cites W1644652583 @default.
- W2116724101 cites W1880262756 @default.
- W2116724101 cites W2031284124 @default.
- W2116724101 cites W2050971845 @default.
- W2116724101 cites W2096610504 @default.
- W2116724101 cites W2099260490 @default.
- W2116724101 cites W2110027950 @default.
- W2116724101 cites W2118714763 @default.
- W2116724101 cites W2127602683 @default.
- W2116724101 cites W2131034123 @default.
- W2116724101 cites W2154594550 @default.
- W2116724101 cites W2159518412 @default.
- W2116724101 cites W2170106220 @default.
- W2116724101 cites W67002875 @default.
- W2116724101 cites W85801758 @default.
- W2116724101 doi "https://doi.org/10.1109/icassp.2012.6288907" @default.
- W2116724101 hasPublicationYear "2012" @default.
- W2116724101 type Work @default.
- W2116724101 sameAs 2116724101 @default.
- W2116724101 citedByCount "1" @default.
- W2116724101 countsByYear W21167241012012 @default.
- W2116724101 crossrefType "proceedings-article" @default.
- W2116724101 hasAuthorship W2116724101A5007695606 @default.
- W2116724101 hasAuthorship W2116724101A5019973376 @default.
- W2116724101 hasConcept C100279451 @default.
- W2116724101 hasConcept C137293760 @default.
- W2116724101 hasConcept C154945302 @default.
- W2116724101 hasConcept C171686336 @default.
- W2116724101 hasConcept C199360897 @default.
- W2116724101 hasConcept C204321447 @default.
- W2116724101 hasConcept C2776214188 @default.
- W2116724101 hasConcept C28490314 @default.
- W2116724101 hasConcept C40969351 @default.
- W2116724101 hasConcept C41008148 @default.
- W2116724101 hasConcept C43521106 @default.
- W2116724101 hasConcept C500882744 @default.
- W2116724101 hasConcept C73555534 @default.
- W2116724101 hasConceptScore W2116724101C100279451 @default.
- W2116724101 hasConceptScore W2116724101C137293760 @default.
- W2116724101 hasConceptScore W2116724101C154945302 @default.
- W2116724101 hasConceptScore W2116724101C171686336 @default.
- W2116724101 hasConceptScore W2116724101C199360897 @default.
- W2116724101 hasConceptScore W2116724101C204321447 @default.
- W2116724101 hasConceptScore W2116724101C2776214188 @default.
- W2116724101 hasConceptScore W2116724101C28490314 @default.
- W2116724101 hasConceptScore W2116724101C40969351 @default.
- W2116724101 hasConceptScore W2116724101C41008148 @default.
- W2116724101 hasConceptScore W2116724101C43521106 @default.
- W2116724101 hasConceptScore W2116724101C500882744 @default.
- W2116724101 hasConceptScore W2116724101C73555534 @default.
- W2116724101 hasLocation W21167241011 @default.
- W2116724101 hasOpenAccess W2116724101 @default.
- W2116724101 hasPrimaryLocation W21167241011 @default.
- W2116724101 hasRelatedWork W142374489 @default.
- W2116724101 hasRelatedWork W1633122917 @default.
- W2116724101 hasRelatedWork W2021343534 @default.
- W2116724101 hasRelatedWork W2116724101 @default.
- W2116724101 hasRelatedWork W2120010536 @default.
- W2116724101 hasRelatedWork W2250993361 @default.
- W2116724101 hasRelatedWork W2399550240 @default.
- W2116724101 hasRelatedWork W2808739814 @default.
- W2116724101 hasRelatedWork W2884815824 @default.
- W2116724101 hasRelatedWork W3035215391 @default.
- W2116724101 isParatext "false" @default.
- W2116724101 isRetracted "false" @default.
- W2116724101 magId "2116724101" @default.
- W2116724101 workType "article" @default.