Matches in SemOpenAlex for { <https://semopenalex.org/work/W3093535103> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W3093535103 abstract "In this work, we introduce metric learning (ML) to enhance the deep embedding learning for text-independent speaker verification (SV). Specifically, the deep speaker embedding network is trained with conventional cross entropy loss and auxiliary pair-based ML loss function. For the auxiliary ML task, training samples of a mini-batch are first arranged into pairs, then positive and negative pairs are selected and weighted through their own and relative similarities, and finally the auxiliary ML loss is calculated by the similarity of the selected pairs. To evaluate the proposed method, we conduct experiments on the Speaker in the Wild (SITW) dataset. The results demonstrate the effectiveness of the proposed method." @default.
- W3093535103 created "2020-10-29" @default.
- W3093535103 creator A5003578886 @default.
- W3093535103 creator A5006734883 @default.
- W3093535103 creator A5023495197 @default.
- W3093535103 creator A5038369085 @default.
- W3093535103 creator A5042426677 @default.
- W3093535103 date "2020-10-21" @default.
- W3093535103 modified "2023-10-16" @default.
- W3093535103 title "Multi-task Metric Learning for Text-independent Speaker Verification" @default.
- W3093535103 cites W1524333225 @default.
- W3093535103 cites W2150769028 @default.
- W3093535103 cites W2194775991 @default.
- W3093535103 cites W2271840356 @default.
- W3093535103 cites W2395750323 @default.
- W3093535103 cites W2404617565 @default.
- W3093535103 cites W2516764878 @default.
- W3093535103 cites W2726515241 @default.
- W3093535103 cites W2746742816 @default.
- W3093535103 cites W2748488820 @default.
- W3093535103 cites W2808631503 @default.
- W3093535103 cites W2889016587 @default.
- W3093535103 cites W2890964092 @default.
- W3093535103 cites W2963386851 @default.
- W3093535103 cites W2963418523 @default.
- W3093535103 cites W2963702081 @default.
- W3093535103 cites W2964271799 @default.
- W3093535103 cites W2964351684 @default.
- W3093535103 cites W2999108204 @default.
- W3093535103 doi "https://doi.org/10.48550/arxiv.2010.10919" @default.
- W3093535103 hasPublicationYear "2020" @default.
- W3093535103 type Work @default.
- W3093535103 sameAs 3093535103 @default.
- W3093535103 citedByCount "0" @default.
- W3093535103 crossrefType "posted-content" @default.
- W3093535103 hasAuthorship W3093535103A5003578886 @default.
- W3093535103 hasAuthorship W3093535103A5006734883 @default.
- W3093535103 hasAuthorship W3093535103A5023495197 @default.
- W3093535103 hasAuthorship W3093535103A5038369085 @default.
- W3093535103 hasAuthorship W3093535103A5042426677 @default.
- W3093535103 hasBestOaLocation W30935351031 @default.
- W3093535103 hasConcept C103278499 @default.
- W3093535103 hasConcept C106301342 @default.
- W3093535103 hasConcept C115961682 @default.
- W3093535103 hasConcept C121332964 @default.
- W3093535103 hasConcept C127413603 @default.
- W3093535103 hasConcept C133892786 @default.
- W3093535103 hasConcept C14036430 @default.
- W3093535103 hasConcept C153180895 @default.
- W3093535103 hasConcept C154945302 @default.
- W3093535103 hasConcept C167981619 @default.
- W3093535103 hasConcept C176217482 @default.
- W3093535103 hasConcept C201995342 @default.
- W3093535103 hasConcept C21547014 @default.
- W3093535103 hasConcept C2780451532 @default.
- W3093535103 hasConcept C28490314 @default.
- W3093535103 hasConcept C2982762665 @default.
- W3093535103 hasConcept C41008148 @default.
- W3093535103 hasConcept C41608201 @default.
- W3093535103 hasConcept C62520636 @default.
- W3093535103 hasConcept C78458016 @default.
- W3093535103 hasConcept C86803240 @default.
- W3093535103 hasConceptScore W3093535103C103278499 @default.
- W3093535103 hasConceptScore W3093535103C106301342 @default.
- W3093535103 hasConceptScore W3093535103C115961682 @default.
- W3093535103 hasConceptScore W3093535103C121332964 @default.
- W3093535103 hasConceptScore W3093535103C127413603 @default.
- W3093535103 hasConceptScore W3093535103C133892786 @default.
- W3093535103 hasConceptScore W3093535103C14036430 @default.
- W3093535103 hasConceptScore W3093535103C153180895 @default.
- W3093535103 hasConceptScore W3093535103C154945302 @default.
- W3093535103 hasConceptScore W3093535103C167981619 @default.
- W3093535103 hasConceptScore W3093535103C176217482 @default.
- W3093535103 hasConceptScore W3093535103C201995342 @default.
- W3093535103 hasConceptScore W3093535103C21547014 @default.
- W3093535103 hasConceptScore W3093535103C2780451532 @default.
- W3093535103 hasConceptScore W3093535103C28490314 @default.
- W3093535103 hasConceptScore W3093535103C2982762665 @default.
- W3093535103 hasConceptScore W3093535103C41008148 @default.
- W3093535103 hasConceptScore W3093535103C41608201 @default.
- W3093535103 hasConceptScore W3093535103C62520636 @default.
- W3093535103 hasConceptScore W3093535103C78458016 @default.
- W3093535103 hasConceptScore W3093535103C86803240 @default.
- W3093535103 hasLocation W30935351031 @default.
- W3093535103 hasOpenAccess W3093535103 @default.
- W3093535103 hasPrimaryLocation W30935351031 @default.
- W3093535103 hasRelatedWork W1982992797 @default.
- W3093535103 hasRelatedWork W2797896074 @default.
- W3093535103 hasRelatedWork W3003582230 @default.
- W3093535103 hasRelatedWork W3015923889 @default.
- W3093535103 hasRelatedWork W3093535103 @default.
- W3093535103 hasRelatedWork W3163059624 @default.
- W3093535103 hasRelatedWork W3179294409 @default.
- W3093535103 hasRelatedWork W4226428303 @default.
- W3093535103 hasRelatedWork W4287102200 @default.
- W3093535103 hasRelatedWork W4313033198 @default.
- W3093535103 isParatext "false" @default.
- W3093535103 isRetracted "false" @default.
- W3093535103 magId "3093535103" @default.
- W3093535103 workType "article" @default.