Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100593158> ?p ?o ?g. }
Showing items 1 to 88 of
88
with 100 items per page.
- W3100593158 endingPage "86" @default.
- W3100593158 startingPage "79" @default.
- W3100593158 abstract "AbstractDeep speaker-embedding neural network trained with a discriminative loss function is widely known to be effective for speaker verification task. Notably, angular margin softmax loss, and its variants, were proposed to promote intra-class compactness. However, it is worth noticing that these methods are not effective enough in enhancing inter-class separability. In this paper, we present a ranked weight loss which explicitly encourages intra-class compactness and enhances inter-class separability simultaneously. During the neural network training process, the most attention is given to the target speaker in order to encourage intra-class compactness. Next, its nearest neighbor who has the greatest impact on the correct classification gets the second most attention while the least attention is paid to its farthest neighbor. Experimental results on VoxCeleb1, CN-Celeb and the Speakers in the Wild (SITW) core-core condition show that the proposed ranked weight loss achieves state-of-the-art performance.KeywordsSpeaker verificationSpeaker embeddingIntra-class compactnessInter-class separability" @default.
- W3100593158 created "2020-11-23" @default.
- W3100593158 creator A5004287909 @default.
- W3100593158 creator A5017251198 @default.
- W3100593158 creator A5050763764 @default.
- W3100593158 creator A5081055225 @default.
- W3100593158 creator A5087592842 @default.
- W3100593158 date "2020-01-01" @default.
- W3100593158 modified "2023-09-26" @default.
- W3100593158 title "Deep Discriminative Embedding with Ranked Weight for Speaker Verification" @default.
- W3100593158 cites W2121812409 @default.
- W3100593158 cites W2150769028 @default.
- W3100593158 cites W2516764878 @default.
- W3100593158 cites W2726515241 @default.
- W3100593158 cites W2808631503 @default.
- W3100593158 cites W2890964092 @default.
- W3100593158 cites W2961224374 @default.
- W3100593158 cites W2963371159 @default.
- W3100593158 cites W2963466847 @default.
- W3100593158 cites W2969985801 @default.
- W3100593158 cites W2972986505 @default.
- W3100593158 cites W3015598461 @default.
- W3100593158 cites W3103152812 @default.
- W3100593158 doi "https://doi.org/10.1007/978-3-030-63823-8_10" @default.
- W3100593158 hasPublicationYear "2020" @default.
- W3100593158 type Work @default.
- W3100593158 sameAs 3100593158 @default.
- W3100593158 citedByCount "1" @default.
- W3100593158 countsByYear W31005931582021 @default.
- W3100593158 crossrefType "book-chapter" @default.
- W3100593158 hasAuthorship W3100593158A5004287909 @default.
- W3100593158 hasAuthorship W3100593158A5017251198 @default.
- W3100593158 hasAuthorship W3100593158A5050763764 @default.
- W3100593158 hasAuthorship W3100593158A5081055225 @default.
- W3100593158 hasAuthorship W3100593158A5087592842 @default.
- W3100593158 hasConcept C119857082 @default.
- W3100593158 hasConcept C14036430 @default.
- W3100593158 hasConcept C153180895 @default.
- W3100593158 hasConcept C154945302 @default.
- W3100593158 hasConcept C18648836 @default.
- W3100593158 hasConcept C188441871 @default.
- W3100593158 hasConcept C202444582 @default.
- W3100593158 hasConcept C2777212361 @default.
- W3100593158 hasConcept C28490314 @default.
- W3100593158 hasConcept C33923547 @default.
- W3100593158 hasConcept C41008148 @default.
- W3100593158 hasConcept C41608201 @default.
- W3100593158 hasConcept C50644808 @default.
- W3100593158 hasConcept C774472 @default.
- W3100593158 hasConcept C78458016 @default.
- W3100593158 hasConcept C86803240 @default.
- W3100593158 hasConcept C97931131 @default.
- W3100593158 hasConceptScore W3100593158C119857082 @default.
- W3100593158 hasConceptScore W3100593158C14036430 @default.
- W3100593158 hasConceptScore W3100593158C153180895 @default.
- W3100593158 hasConceptScore W3100593158C154945302 @default.
- W3100593158 hasConceptScore W3100593158C18648836 @default.
- W3100593158 hasConceptScore W3100593158C188441871 @default.
- W3100593158 hasConceptScore W3100593158C202444582 @default.
- W3100593158 hasConceptScore W3100593158C2777212361 @default.
- W3100593158 hasConceptScore W3100593158C28490314 @default.
- W3100593158 hasConceptScore W3100593158C33923547 @default.
- W3100593158 hasConceptScore W3100593158C41008148 @default.
- W3100593158 hasConceptScore W3100593158C41608201 @default.
- W3100593158 hasConceptScore W3100593158C50644808 @default.
- W3100593158 hasConceptScore W3100593158C774472 @default.
- W3100593158 hasConceptScore W3100593158C78458016 @default.
- W3100593158 hasConceptScore W3100593158C86803240 @default.
- W3100593158 hasConceptScore W3100593158C97931131 @default.
- W3100593158 hasLocation W31005931581 @default.
- W3100593158 hasOpenAccess W3100593158 @default.
- W3100593158 hasPrimaryLocation W31005931581 @default.
- W3100593158 hasRelatedWork W2743258233 @default.
- W3100593158 hasRelatedWork W2944681685 @default.
- W3100593158 hasRelatedWork W2951495199 @default.
- W3100593158 hasRelatedWork W2966628364 @default.
- W3100593158 hasRelatedWork W3095203785 @default.
- W3100593158 hasRelatedWork W3100593158 @default.
- W3100593158 hasRelatedWork W4214501853 @default.
- W3100593158 hasRelatedWork W4214883769 @default.
- W3100593158 hasRelatedWork W4307536017 @default.
- W3100593158 hasRelatedWork W2890834819 @default.
- W3100593158 isParatext "false" @default.
- W3100593158 isRetracted "false" @default.
- W3100593158 magId "3100593158" @default.
- W3100593158 workType "book-chapter" @default.