FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4289134328> ?p ?o ?g. }

• W4289134328 endingPage "1046" @default.
• W4289134328 startingPage "1046" @default.
• W4289134328 abstract "Cross-corpus speech emotion recognition (SER) is a challenging task, and its difficulty lies in the mismatch between the feature distributions of the training (source domain) and testing (target domain) data, leading to the performance degradation when the model deals with new domain data. Previous works explore utilizing domain adaptation (DA) to eliminate the domain shift between the source and target domains and have achieved the promising performance in SER. However, these methods mainly treat cross-corpus tasks simply as the DA problem, directly aligning the distributions across domains in a common feature space. In this case, excessively narrowing the domain distance will impair the emotion discrimination of speech features since it is difficult to maintain the completeness of the emotion space only by an emotion classifier. To overcome this issue, we propose a progressively discriminative transfer network (PDTN) for cross-corpus SER in this paper, which can enhance the emotion discrimination ability of speech features while eliminating the mismatch between the source and target corpora. In detail, we design two special losses in the feature layers of PDTN, i.e., emotion discriminant loss Ld and distribution alignment loss La. By incorporating prior knowledge of speech emotion into feature learning (i.e., high and low valence speech emotion features have their respective cluster centers), we integrate a valence-aware center loss Lv and an emotion-aware center loss Lc as the Ld to guarantee the discriminative learning of speech emotions except an emotion classifier. Furthermore, a multi-layer distribution alignment loss La is adopted to more precisely eliminate the discrepancy of feature distributions between the source and target domains. Finally, through the optimization of PDTN by combining three losses, i.e., cross-entropy loss Le, Ld, and La, we can gradually eliminate the domain mismatch between the source and target corpora while maintaining the emotion discrimination of speech features. Extensive experimental results of six cross-corpus tasks on three datasets, i.e., Emo-DB, eNTERFACE, and CASIA, reveal that our proposed PDTN outperforms the state-of-the-art methods." @default.
• W4289134328 created "2022-08-01" @default.
• W4289134328 creator A5027316177 @default.
• W4289134328 creator A5038686056 @default.
• W4289134328 creator A5054796879 @default.
• W4289134328 creator A5059266417 @default.
• W4289134328 date "2022-07-29" @default.
• W4289134328 modified "2023-10-17" @default.
• W4289134328 title "Progressively Discriminative Transfer Network for Cross-Corpus Speech Emotion Recognition" @default.
• W4289134328 cites W175750906 @default.
• W4289134328 cites W2032254851 @default.
• W4289134328 cites W2113087918 @default.
• W4289134328 cites W2125462608 @default.
• W4289134328 cites W2149628368 @default.
• W4289134328 cites W2168692779 @default.
• W4289134328 cites W2322208196 @default.
• W4289134328 cites W2503045345 @default.
• W4289134328 cites W2614874155 @default.
• W4289134328 cites W2735297260 @default.
• W4289134328 cites W2895006884 @default.
• W4289134328 cites W2963447013 @default.
• W4289134328 cites W2980393359 @default.
• W4289134328 cites W2983165789 @default.
• W4289134328 cites W3021632667 @default.
• W4289134328 cites W3208366642 @default.
• W4289134328 cites W4286544676 @default.
• W4289134328 cites W3161933272 @default.
• W4289134328 doi "https://doi.org/10.3390/e24081046" @default.
• W4289134328 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36010710" @default.
• W4289134328 hasPublicationYear "2022" @default.
• W4289134328 type Work @default.
• W4289134328 citedByCount "4" @default.
• W4289134328 countsByYear W42891343282022 @default.
• W4289134328 countsByYear W42891343282023 @default.
• W4289134328 crossrefType "journal-article" @default.
• W4289134328 hasAuthorship W4289134328A5027316177 @default.
• W4289134328 hasAuthorship W4289134328A5038686056 @default.
• W4289134328 hasAuthorship W4289134328A5054796879 @default.
• W4289134328 hasAuthorship W4289134328A5059266417 @default.
• W4289134328 hasBestOaLocation W42891343281 @default.
• W4289134328 hasConcept C121332964 @default.
• W4289134328 hasConcept C138885662 @default.
• W4289134328 hasConcept C150899416 @default.
• W4289134328 hasConcept C153180895 @default.
• W4289134328 hasConcept C154945302 @default.
• W4289134328 hasConcept C168900304 @default.
• W4289134328 hasConcept C206310091 @default.
• W4289134328 hasConcept C2776401178 @default.
• W4289134328 hasConcept C28490314 @default.
• W4289134328 hasConcept C41008148 @default.
• W4289134328 hasConcept C41895202 @default.
• W4289134328 hasConcept C62520636 @default.
• W4289134328 hasConcept C83665646 @default.
• W4289134328 hasConcept C95623464 @default.
• W4289134328 hasConcept C97931131 @default.
• W4289134328 hasConceptScore W4289134328C121332964 @default.
• W4289134328 hasConceptScore W4289134328C138885662 @default.
• W4289134328 hasConceptScore W4289134328C150899416 @default.
• W4289134328 hasConceptScore W4289134328C153180895 @default.
• W4289134328 hasConceptScore W4289134328C154945302 @default.
• W4289134328 hasConceptScore W4289134328C168900304 @default.
• W4289134328 hasConceptScore W4289134328C206310091 @default.
• W4289134328 hasConceptScore W4289134328C2776401178 @default.
• W4289134328 hasConceptScore W4289134328C28490314 @default.
• W4289134328 hasConceptScore W4289134328C41008148 @default.
• W4289134328 hasConceptScore W4289134328C41895202 @default.
• W4289134328 hasConceptScore W4289134328C62520636 @default.
• W4289134328 hasConceptScore W4289134328C83665646 @default.
• W4289134328 hasConceptScore W4289134328C95623464 @default.
• W4289134328 hasConceptScore W4289134328C97931131 @default.
• W4289134328 hasIssue "8" @default.
• W4289134328 hasLocation W42891343281 @default.
• W4289134328 hasLocation W42891343282 @default.
• W4289134328 hasLocation W42891343283 @default.
• W4289134328 hasLocation W42891343284 @default.
• W4289134328 hasOpenAccess W4289134328 @default.
• W4289134328 hasPrimaryLocation W42891343281 @default.
• W4289134328 hasRelatedWork W1482209366 @default.
• W4289134328 hasRelatedWork W2110523656 @default.
• W4289134328 hasRelatedWork W2153315159 @default.
• W4289134328 hasRelatedWork W259157601 @default.
• W4289134328 hasRelatedWork W2761785940 @default.
• W4289134328 hasRelatedWork W2903290484 @default.
• W4289134328 hasRelatedWork W2965546495 @default.
• W4289134328 hasRelatedWork W2967180365 @default.
• W4289134328 hasRelatedWork W3103844505 @default.
• W4289134328 hasRelatedWork W4205463238 @default.
• W4289134328 hasVolume "24" @default.
• W4289134328 isParatext "false" @default.
• W4289134328 isRetracted "false" @default.
• W4289134328 workType "article" @default.