Matches in SemOpenAlex for { <https://semopenalex.org/work/W4225319488> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4225319488 abstract "The lack of speech data annotated with labels required for spoken language understanding (SLU) is often a major hurdle in building end-to-end (E2E) systems that can directly process speech inputs. In contrast, large amounts of text data with suitable labels are usually available. In this paper, we propose a novel text representation and training methodology that allows E2E SLU systems to be effectively constructed using these text resources. With very limited amounts of additional speech, we show that these models can be further improved to perform at levels close to similar systems built on the full speech datasets. The efficacy of our proposed approach is demonstrated on both intent and entity tasks using three different SLU datasets. With text-only training, the proposed system achieves up to 90% of the performance possible with full speech training. With just an additional 10% of speech data, these models significantly improve further to 97% of full performance." @default.
- W4225319488 created "2022-05-05" @default.
- W4225319488 creator A5003725957 @default.
- W4225319488 creator A5056869363 @default.
- W4225319488 creator A5072442182 @default.
- W4225319488 creator A5079994647 @default.
- W4225319488 date "2022-05-23" @default.
- W4225319488 modified "2023-09-26" @default.
- W4225319488 title "Towards Reducing the Need for Speech Training Data to Build Spoken Language Understanding Systems" @default.
- W4225319488 cites W2077302143 @default.
- W4225319488 cites W2153338568 @default.
- W4225319488 cites W2786839803 @default.
- W4225319488 cites W2891229414 @default.
- W4225319488 cites W2894164357 @default.
- W4225319488 cites W2914417638 @default.
- W4225319488 cites W2936774411 @default.
- W4225319488 cites W2937780860 @default.
- W4225319488 cites W2962760690 @default.
- W4225319488 cites W2963288440 @default.
- W4225319488 cites W2972327934 @default.
- W4225319488 cites W2972584841 @default.
- W4225319488 cites W2973127116 @default.
- W4225319488 cites W3007227084 @default.
- W4225319488 cites W3015686596 @default.
- W4225319488 cites W3015797571 @default.
- W4225319488 cites W3015885816 @default.
- W4225319488 cites W3016006013 @default.
- W4225319488 cites W3016262400 @default.
- W4225319488 cites W3095552229 @default.
- W4225319488 cites W3096251052 @default.
- W4225319488 cites W3097964672 @default.
- W4225319488 cites W3161000786 @default.
- W4225319488 cites W3163300396 @default.
- W4225319488 doi "https://doi.org/10.1109/icassp43922.2022.9747555" @default.
- W4225319488 hasPublicationYear "2022" @default.
- W4225319488 type Work @default.
- W4225319488 citedByCount "3" @default.
- W4225319488 countsByYear W42253194882023 @default.
- W4225319488 crossrefType "proceedings-article" @default.
- W4225319488 hasAuthorship W4225319488A5003725957 @default.
- W4225319488 hasAuthorship W4225319488A5056869363 @default.
- W4225319488 hasAuthorship W4225319488A5072442182 @default.
- W4225319488 hasAuthorship W4225319488A5079994647 @default.
- W4225319488 hasBestOaLocation W42253194882 @default.
- W4225319488 hasConcept C111919701 @default.
- W4225319488 hasConcept C121332964 @default.
- W4225319488 hasConcept C137293760 @default.
- W4225319488 hasConcept C153294291 @default.
- W4225319488 hasConcept C154945302 @default.
- W4225319488 hasConcept C17744445 @default.
- W4225319488 hasConcept C199539241 @default.
- W4225319488 hasConcept C204321447 @default.
- W4225319488 hasConcept C2776230583 @default.
- W4225319488 hasConcept C2776359362 @default.
- W4225319488 hasConcept C2777211547 @default.
- W4225319488 hasConcept C28490314 @default.
- W4225319488 hasConcept C41008148 @default.
- W4225319488 hasConcept C51632099 @default.
- W4225319488 hasConcept C61328038 @default.
- W4225319488 hasConcept C94625758 @default.
- W4225319488 hasConcept C98045186 @default.
- W4225319488 hasConceptScore W4225319488C111919701 @default.
- W4225319488 hasConceptScore W4225319488C121332964 @default.
- W4225319488 hasConceptScore W4225319488C137293760 @default.
- W4225319488 hasConceptScore W4225319488C153294291 @default.
- W4225319488 hasConceptScore W4225319488C154945302 @default.
- W4225319488 hasConceptScore W4225319488C17744445 @default.
- W4225319488 hasConceptScore W4225319488C199539241 @default.
- W4225319488 hasConceptScore W4225319488C204321447 @default.
- W4225319488 hasConceptScore W4225319488C2776230583 @default.
- W4225319488 hasConceptScore W4225319488C2776359362 @default.
- W4225319488 hasConceptScore W4225319488C2777211547 @default.
- W4225319488 hasConceptScore W4225319488C28490314 @default.
- W4225319488 hasConceptScore W4225319488C41008148 @default.
- W4225319488 hasConceptScore W4225319488C51632099 @default.
- W4225319488 hasConceptScore W4225319488C61328038 @default.
- W4225319488 hasConceptScore W4225319488C94625758 @default.
- W4225319488 hasConceptScore W4225319488C98045186 @default.
- W4225319488 hasLocation W42253194881 @default.
- W4225319488 hasLocation W42253194882 @default.
- W4225319488 hasOpenAccess W4225319488 @default.
- W4225319488 hasPrimaryLocation W42253194881 @default.
- W4225319488 hasRelatedWork W14090282 @default.
- W4225319488 hasRelatedWork W14230040 @default.
- W4225319488 hasRelatedWork W14471622 @default.
- W4225319488 hasRelatedWork W1614943 @default.
- W4225319488 hasRelatedWork W3939953 @default.
- W4225319488 hasRelatedWork W404804 @default.
- W4225319488 hasRelatedWork W5871360 @default.
- W4225319488 hasRelatedWork W7655147 @default.
- W4225319488 hasRelatedWork W9280962 @default.
- W4225319488 hasRelatedWork W9840409 @default.
- W4225319488 isParatext "false" @default.
- W4225319488 isRetracted "false" @default.
- W4225319488 workType "article" @default.