Matches in SemOpenAlex for { <https://semopenalex.org/work/W2951193022> ?p ?o ?g. }
- W2951193022 abstract "End-to-end (E2E) models have been explored for large speech corpora and have been found to match or outperform traditional pipeline-based systems in some languages. However, most prior work on end-to-end models use speech corpora exceeding hundreds or thousands of hours. In this study, we explore end-to-end models for code-switched Hindi-English language with less than 50 hours of data. We utilize two specific measures to improve network performance in the low-resource setting, namely multi-task learning (MTL) and balancing the corpus to deal with the inherent class imbalance problem i.e. the skewed frequency distribution over graphemes. We compare the results of the proposed approaches with traditional, cascaded ASR systems. While the lack of data adversely affects the performance of end-to-end models, we see promising improvements with MTL and balancing the corpus." @default.
- W2951193022 created "2019-06-27" @default.
- W2951193022 creator A5005220677 @default.
- W2951193022 creator A5005513786 @default.
- W2951193022 creator A5036738038 @default.
- W2951193022 creator A5042205423 @default.
- W2951193022 creator A5061176524 @default.
- W2951193022 date "2019-06-22" @default.
- W2951193022 modified "2023-09-27" @default.
- W2951193022 title "End-to-End ASR for Code-switched Hindi-English Speech." @default.
- W2951193022 cites W1524333225 @default.
- W2951193022 cites W1586532344 @default.
- W2951193022 cites W2102113734 @default.
- W2951193022 cites W2119191234 @default.
- W2951193022 cites W2127141656 @default.
- W2951193022 cites W2327501763 @default.
- W2951193022 cites W2520160253 @default.
- W2951193022 cites W2526425061 @default.
- W2951193022 cites W2530876040 @default.
- W2951193022 cites W2766219058 @default.
- W2951193022 cites W2767106145 @default.
- W2951193022 cites W2799800213 @default.
- W2951193022 cites W2809978471 @default.
- W2951193022 cites W2889163603 @default.
- W2951193022 cites W2899073901 @default.
- W2951193022 cites W2899299209 @default.
- W2951193022 cites W2962780374 @default.
- W2951193022 cites W2962824709 @default.
- W2951193022 cites W2962826786 @default.
- W2951193022 cites W2962893195 @default.
- W2951193022 cites W2963071736 @default.
- W2951193022 cites W2963211739 @default.
- W2951193022 cites W2963403664 @default.
- W2951193022 cites W2964309797 @default.
- W2951193022 cites W6908809 @default.
- W2951193022 cites W854541894 @default.
- W2951193022 hasPublicationYear "2019" @default.
- W2951193022 type Work @default.
- W2951193022 sameAs 2951193022 @default.
- W2951193022 citedByCount "1" @default.
- W2951193022 countsByYear W29511930222020 @default.
- W2951193022 crossrefType "posted-content" @default.
- W2951193022 hasAuthorship W2951193022A5005220677 @default.
- W2951193022 hasAuthorship W2951193022A5005513786 @default.
- W2951193022 hasAuthorship W2951193022A5036738038 @default.
- W2951193022 hasAuthorship W2951193022A5042205423 @default.
- W2951193022 hasAuthorship W2951193022A5061176524 @default.
- W2951193022 hasConcept C137293760 @default.
- W2951193022 hasConcept C154945302 @default.
- W2951193022 hasConcept C162324750 @default.
- W2951193022 hasConcept C177264268 @default.
- W2951193022 hasConcept C187736073 @default.
- W2951193022 hasConcept C199360897 @default.
- W2951193022 hasConcept C204321447 @default.
- W2951193022 hasConcept C2776760102 @default.
- W2951193022 hasConcept C2780451532 @default.
- W2951193022 hasConcept C28490314 @default.
- W2951193022 hasConcept C41008148 @default.
- W2951193022 hasConcept C43521106 @default.
- W2951193022 hasConcept C519982507 @default.
- W2951193022 hasConcept C74296488 @default.
- W2951193022 hasConceptScore W2951193022C137293760 @default.
- W2951193022 hasConceptScore W2951193022C154945302 @default.
- W2951193022 hasConceptScore W2951193022C162324750 @default.
- W2951193022 hasConceptScore W2951193022C177264268 @default.
- W2951193022 hasConceptScore W2951193022C187736073 @default.
- W2951193022 hasConceptScore W2951193022C199360897 @default.
- W2951193022 hasConceptScore W2951193022C204321447 @default.
- W2951193022 hasConceptScore W2951193022C2776760102 @default.
- W2951193022 hasConceptScore W2951193022C2780451532 @default.
- W2951193022 hasConceptScore W2951193022C28490314 @default.
- W2951193022 hasConceptScore W2951193022C41008148 @default.
- W2951193022 hasConceptScore W2951193022C43521106 @default.
- W2951193022 hasConceptScore W2951193022C519982507 @default.
- W2951193022 hasConceptScore W2951193022C74296488 @default.
- W2951193022 hasLocation W29511930221 @default.
- W2951193022 hasOpenAccess W2951193022 @default.
- W2951193022 hasPrimaryLocation W29511930221 @default.
- W2951193022 hasRelatedWork W2405083174 @default.
- W2951193022 hasRelatedWork W2939757332 @default.
- W2951193022 hasRelatedWork W2964014801 @default.
- W2951193022 hasRelatedWork W2972320704 @default.
- W2951193022 hasRelatedWork W2973227840 @default.
- W2951193022 hasRelatedWork W2980300246 @default.
- W2951193022 hasRelatedWork W2995732811 @default.
- W2951193022 hasRelatedWork W3007376164 @default.
- W2951193022 hasRelatedWork W3012101946 @default.
- W2951193022 hasRelatedWork W3029522047 @default.
- W2951193022 hasRelatedWork W3034335198 @default.
- W2951193022 hasRelatedWork W3035490255 @default.
- W2951193022 hasRelatedWork W3038127439 @default.
- W2951193022 hasRelatedWork W3040454670 @default.
- W2951193022 hasRelatedWork W3042007685 @default.
- W2951193022 hasRelatedWork W3093964916 @default.
- W2951193022 hasRelatedWork W3161127957 @default.
- W2951193022 hasRelatedWork W3162345085 @default.
- W2951193022 hasRelatedWork W3180315287 @default.
- W2951193022 hasRelatedWork W3197472475 @default.
- W2951193022 isParatext "false" @default.
- W2951193022 isRetracted "false" @default.