Matches in SemOpenAlex for { <https://semopenalex.org/work/W3200768929> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3200768929 abstract "End-to-end speech recognition systems usually require huge amounts of labeling resource, while annotating the speech data is complicated and expensive. Active learning is the solution by selecting the most valuable samples for annotation. In this paper, we proposed to use a predicted loss that estimates the uncertainty of the sample. The CTC (Connectionist Temporal Classification) and attention loss are informative for speech recognition since they are computed based on all decoding paths and alignments. We defined an end-to-end active learning pipeline, training an ASR/LP (Automatic Speech Recognition/Loss Prediction) joint model. The proposed approach was validated on an English and a Chinese speech recognition task. The experiments show that our approach achieves competitive results, outperforming random selection, least confidence, and estimated loss method." @default.
- W3200768929 created "2021-09-27" @default.
- W3200768929 creator A5016038454 @default.
- W3200768929 creator A5074472751 @default.
- W3200768929 creator A5076982760 @default.
- W3200768929 creator A5084554132 @default.
- W3200768929 date "2021-07-18" @default.
- W3200768929 modified "2023-09-30" @default.
- W3200768929 title "Loss Prediction: End-to-End Active Learning Approach For Speech Recognition" @default.
- W3200768929 cites W1513874326 @default.
- W3200768929 cites W1536680647 @default.
- W3200768929 cites W1978633512 @default.
- W3200768929 cites W2124244761 @default.
- W3200768929 cites W2124415900 @default.
- W3200768929 cites W2127141656 @default.
- W3200768929 cites W2127499922 @default.
- W3200768929 cites W2171671120 @default.
- W3200768929 cites W2471138382 @default.
- W3200768929 cites W2526425061 @default.
- W3200768929 cites W2766219058 @default.
- W3200768929 cites W2892009249 @default.
- W3200768929 cites W2935875038 @default.
- W3200768929 cites W2951911250 @default.
- W3200768929 cites W2956371155 @default.
- W3200768929 cites W2962780374 @default.
- W3200768929 cites W2963015987 @default.
- W3200768929 cites W2963242190 @default.
- W3200768929 cites W2972551158 @default.
- W3200768929 cites W2986514296 @default.
- W3200768929 cites W2989227218 @default.
- W3200768929 cites W3089603314 @default.
- W3200768929 cites W3090241307 @default.
- W3200768929 cites W3149836819 @default.
- W3200768929 cites W4234016251 @default.
- W3200768929 doi "https://doi.org/10.1109/ijcnn52387.2021.9533839" @default.
- W3200768929 hasPublicationYear "2021" @default.
- W3200768929 type Work @default.
- W3200768929 sameAs 3200768929 @default.
- W3200768929 citedByCount "3" @default.
- W3200768929 countsByYear W32007689292022 @default.
- W3200768929 countsByYear W32007689292023 @default.
- W3200768929 crossrefType "proceedings-article" @default.
- W3200768929 hasAuthorship W3200768929A5016038454 @default.
- W3200768929 hasAuthorship W3200768929A5074472751 @default.
- W3200768929 hasAuthorship W3200768929A5076982760 @default.
- W3200768929 hasAuthorship W3200768929A5084554132 @default.
- W3200768929 hasBestOaLocation W32007689292 @default.
- W3200768929 hasConcept C119857082 @default.
- W3200768929 hasConcept C153180895 @default.
- W3200768929 hasConcept C154945302 @default.
- W3200768929 hasConcept C162324750 @default.
- W3200768929 hasConcept C187736073 @default.
- W3200768929 hasConcept C199360897 @default.
- W3200768929 hasConcept C2776321320 @default.
- W3200768929 hasConcept C2780451532 @default.
- W3200768929 hasConcept C28490314 @default.
- W3200768929 hasConcept C41008148 @default.
- W3200768929 hasConcept C43521106 @default.
- W3200768929 hasConcept C50644808 @default.
- W3200768929 hasConcept C57273362 @default.
- W3200768929 hasConcept C74296488 @default.
- W3200768929 hasConcept C76155785 @default.
- W3200768929 hasConcept C8521452 @default.
- W3200768929 hasConceptScore W3200768929C119857082 @default.
- W3200768929 hasConceptScore W3200768929C153180895 @default.
- W3200768929 hasConceptScore W3200768929C154945302 @default.
- W3200768929 hasConceptScore W3200768929C162324750 @default.
- W3200768929 hasConceptScore W3200768929C187736073 @default.
- W3200768929 hasConceptScore W3200768929C199360897 @default.
- W3200768929 hasConceptScore W3200768929C2776321320 @default.
- W3200768929 hasConceptScore W3200768929C2780451532 @default.
- W3200768929 hasConceptScore W3200768929C28490314 @default.
- W3200768929 hasConceptScore W3200768929C41008148 @default.
- W3200768929 hasConceptScore W3200768929C43521106 @default.
- W3200768929 hasConceptScore W3200768929C50644808 @default.
- W3200768929 hasConceptScore W3200768929C57273362 @default.
- W3200768929 hasConceptScore W3200768929C74296488 @default.
- W3200768929 hasConceptScore W3200768929C76155785 @default.
- W3200768929 hasConceptScore W3200768929C8521452 @default.
- W3200768929 hasLocation W32007689291 @default.
- W3200768929 hasLocation W32007689292 @default.
- W3200768929 hasOpenAccess W3200768929 @default.
- W3200768929 hasPrimaryLocation W32007689291 @default.
- W3200768929 hasRelatedWork W1736701665 @default.
- W3200768929 hasRelatedWork W2592356860 @default.
- W3200768929 hasRelatedWork W2782005958 @default.
- W3200768929 hasRelatedWork W2941320171 @default.
- W3200768929 hasRelatedWork W2963211739 @default.
- W3200768929 hasRelatedWork W2963414781 @default.
- W3200768929 hasRelatedWork W3112480982 @default.
- W3200768929 hasRelatedWork W3193318782 @default.
- W3200768929 hasRelatedWork W3196238820 @default.
- W3200768929 hasRelatedWork W4322774807 @default.
- W3200768929 isParatext "false" @default.
- W3200768929 isRetracted "false" @default.
- W3200768929 magId "3200768929" @default.
- W3200768929 workType "article" @default.