Matches in SemOpenAlex for { <https://semopenalex.org/work/W2894690744> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W2894690744 abstract "For many languages in the world, not enough (annotated) speech data is available to train an ASR system. Recently, we proposed a cross-language method for training an ASR system using linguistic knowledge and semi-supervised training. Here, we apply this approach to the low-resource language Mboshi. Using an ASR system trained on Dutch, Mboshi acoustic units were first created using cross-language initialization of the phoneme vectors in the output layer. Subsequently, this adapted system was retrained using Mboshi self-labels. Two training methods were investigated: retraining of only the output layer and retraining the full deep neural network (DNN). The resulting Mboshi system was analyzed by investigating per phoneme accuracies, phoneme confusions, and by visualizing the hidden layers of the DNNs prior to and following retraining with the self-labels. Results showed a fairly similar performance for the two training methods but a better phoneme representation for the fully retrained DNN." @default.
- W2894690744 created "2018-10-12" @default.
- W2894690744 creator A5004778663 @default.
- W2894690744 creator A5048860892 @default.
- W2894690744 creator A5050632169 @default.
- W2894690744 creator A5060929395 @default.
- W2894690744 date "2018-08-29" @default.
- W2894690744 modified "2023-09-30" @default.
- W2894690744 title "Building an ASR System for Mboshi Using A Cross-Language Definition of Acoustic Units Approach" @default.
- W2894690744 cites W1520968739 @default.
- W2894690744 cites W1524333225 @default.
- W2894690744 cites W2025482506 @default.
- W2894690744 cites W2099621636 @default.
- W2894690744 cites W2160003195 @default.
- W2894690744 cites W2271840356 @default.
- W2894690744 cites W2307960051 @default.
- W2894690744 cites W2347098582 @default.
- W2894690744 cites W2347145335 @default.
- W2894690744 cites W2545319977 @default.
- W2894690744 cites W2747287964 @default.
- W2894690744 cites W2889500720 @default.
- W2894690744 cites W2962862718 @default.
- W2894690744 doi "https://doi.org/10.21437/sltu.2018-35" @default.
- W2894690744 hasPublicationYear "2018" @default.
- W2894690744 type Work @default.
- W2894690744 sameAs 2894690744 @default.
- W2894690744 citedByCount "4" @default.
- W2894690744 countsByYear W28946907442019 @default.
- W2894690744 countsByYear W28946907442020 @default.
- W2894690744 crossrefType "proceedings-article" @default.
- W2894690744 hasAuthorship W2894690744A5004778663 @default.
- W2894690744 hasAuthorship W2894690744A5048860892 @default.
- W2894690744 hasAuthorship W2894690744A5050632169 @default.
- W2894690744 hasAuthorship W2894690744A5060929395 @default.
- W2894690744 hasBestOaLocation W28946907442 @default.
- W2894690744 hasConcept C114466953 @default.
- W2894690744 hasConcept C121332964 @default.
- W2894690744 hasConcept C137293760 @default.
- W2894690744 hasConcept C144133560 @default.
- W2894690744 hasConcept C153294291 @default.
- W2894690744 hasConcept C154945302 @default.
- W2894690744 hasConcept C155202549 @default.
- W2894690744 hasConcept C17744445 @default.
- W2894690744 hasConcept C178790620 @default.
- W2894690744 hasConcept C185592680 @default.
- W2894690744 hasConcept C199360897 @default.
- W2894690744 hasConcept C199539241 @default.
- W2894690744 hasConcept C204321447 @default.
- W2894690744 hasConcept C2776359362 @default.
- W2894690744 hasConcept C2777211547 @default.
- W2894690744 hasConcept C2778712577 @default.
- W2894690744 hasConcept C2779227376 @default.
- W2894690744 hasConcept C28490314 @default.
- W2894690744 hasConcept C41008148 @default.
- W2894690744 hasConcept C50644808 @default.
- W2894690744 hasConcept C51632099 @default.
- W2894690744 hasConcept C94625758 @default.
- W2894690744 hasConceptScore W2894690744C114466953 @default.
- W2894690744 hasConceptScore W2894690744C121332964 @default.
- W2894690744 hasConceptScore W2894690744C137293760 @default.
- W2894690744 hasConceptScore W2894690744C144133560 @default.
- W2894690744 hasConceptScore W2894690744C153294291 @default.
- W2894690744 hasConceptScore W2894690744C154945302 @default.
- W2894690744 hasConceptScore W2894690744C155202549 @default.
- W2894690744 hasConceptScore W2894690744C17744445 @default.
- W2894690744 hasConceptScore W2894690744C178790620 @default.
- W2894690744 hasConceptScore W2894690744C185592680 @default.
- W2894690744 hasConceptScore W2894690744C199360897 @default.
- W2894690744 hasConceptScore W2894690744C199539241 @default.
- W2894690744 hasConceptScore W2894690744C204321447 @default.
- W2894690744 hasConceptScore W2894690744C2776359362 @default.
- W2894690744 hasConceptScore W2894690744C2777211547 @default.
- W2894690744 hasConceptScore W2894690744C2778712577 @default.
- W2894690744 hasConceptScore W2894690744C2779227376 @default.
- W2894690744 hasConceptScore W2894690744C28490314 @default.
- W2894690744 hasConceptScore W2894690744C41008148 @default.
- W2894690744 hasConceptScore W2894690744C50644808 @default.
- W2894690744 hasConceptScore W2894690744C51632099 @default.
- W2894690744 hasConceptScore W2894690744C94625758 @default.
- W2894690744 hasLocation W28946907441 @default.
- W2894690744 hasLocation W28946907442 @default.
- W2894690744 hasOpenAccess W2894690744 @default.
- W2894690744 hasPrimaryLocation W28946907441 @default.
- W2894690744 hasRelatedWork W1973340964 @default.
- W2894690744 hasRelatedWork W2082122846 @default.
- W2894690744 hasRelatedWork W2159746348 @default.
- W2894690744 hasRelatedWork W2399715921 @default.
- W2894690744 hasRelatedWork W2753086417 @default.
- W2894690744 hasRelatedWork W2755192592 @default.
- W2894690744 hasRelatedWork W2888315395 @default.
- W2894690744 hasRelatedWork W2894690744 @default.
- W2894690744 hasRelatedWork W2916986993 @default.
- W2894690744 hasRelatedWork W3107474891 @default.
- W2894690744 isParatext "false" @default.
- W2894690744 isRetracted "false" @default.
- W2894690744 magId "2894690744" @default.
- W2894690744 workType "article" @default.