FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3217778739> ?p ?o ?g. }

• W3217778739 endingPage "75" @default.
• W3217778739 startingPage "60" @default.
• W3217778739 abstract "Adapting speaker recognition systems to new environments is a widely-used technique to improve a well-performing model learned from large-scale data towards a task-specific small-scale data scenarios. However, previous studies focus on single domain adaptation, which neglects a more practical scenario where training data are collected from multiple acoustic domains needed in forensic scenarios. Audio analysis for forensic speaker recognition offers unique challenges in model training with multi-domain training data due to location/scenario uncertainty and diversity mismatch between reference and naturalistic field recordings. It is also difficult to directly employ small-scale domain-specific data to train complex neural network architectures due to domain mismatch and performance loss. Fine-tuning is a commonly-used method for adaptation in order to retrain the model with weights initialized from a well-trained model. Alternatively, in this study, three novel adaptation methods based on domain adversarial training, discrepancy minimization, and moment-matching approaches are proposed to further promote adaptation performance across multiple acoustic domains. A comprehensive set of experiments are conducted to demonstrate that: 1) diverse acoustic environments do impact speaker recognition performance, which could advance research in audio forensics, 2) domain adversarial training learns the discriminative features which are also invariant to shifts between domains, 3) discrepancy-minimizing adaptation achieves effective performance simultaneously across multiple acoustic domains, and 4) moment-matching adaptation along with dynamic distribution alignment also significantly promotes speaker recognition performance on each domain, especially for the LENA-field domain with noise compared to all other systems." @default.
• W3217778739 created "2021-12-06" @default.
• W3217778739 creator A5017777141 @default.
• W3217778739 creator A5057910370 @default.
• W3217778739 date "2022-01-01" @default.
• W3217778739 modified "2023-09-30" @default.
• W3217778739 title "Multi-Source Domain Adaptation for Text-Independent Forensic Speaker Recognition" @default.
• W3217778739 cites W1481604723 @default.
• W3217778739 cites W1589137271 @default.
• W3217778739 cites W1936725236 @default.
• W3217778739 cites W2016770751 @default.
• W3217778739 cites W2023238506 @default.
• W3217778739 cites W2101556109 @default.
• W3217778739 cites W2121812409 @default.
• W3217778739 cites W2121892718 @default.
• W3217778739 cites W2124783762 @default.
• W3217778739 cites W2141373863 @default.
• W3217778739 cites W2150769028 @default.
• W3217778739 cites W2293442930 @default.
• W3217778739 cites W2398776621 @default.
• W3217778739 cites W2402146185 @default.
• W3217778739 cites W2408021097 @default.
• W3217778739 cites W2508555431 @default.
• W3217778739 cites W2587150483 @default.
• W3217778739 cites W2593768305 @default.
• W3217778739 cites W2748488820 @default.
• W3217778739 cites W2802973008 @default.
• W3217778739 cites W2808631503 @default.
• W3217778739 cites W2890964092 @default.
• W3217778739 cites W2904209271 @default.
• W3217778739 cites W2936780106 @default.
• W3217778739 cites W2962687275 @default.
• W3217778739 cites W2962788262 @default.
• W3217778739 cites W2962793481 @default.
• W3217778739 cites W2963371159 @default.
• W3217778739 cites W2963506806 @default.
• W3217778739 cites W2968584961 @default.
• W3217778739 cites W2981720610 @default.
• W3217778739 cites W3015964387 @default.
• W3217778739 cites W3096127734 @default.
• W3217778739 cites W3097470351 @default.
• W3217778739 cites W3161606033 @default.
• W3217778739 cites W372778829 @default.
• W3217778739 cites W4234330420 @default.
• W3217778739 cites W4243070457 @default.
• W3217778739 doi "https://doi.org/10.1109/taslp.2021.3130975" @default.
• W3217778739 hasPublicationYear "2022" @default.
• W3217778739 type Work @default.
• W3217778739 sameAs 3217778739 @default.
• W3217778739 citedByCount "8" @default.
• W3217778739 countsByYear W32177787392022 @default.
• W3217778739 countsByYear W32177787392023 @default.
• W3217778739 crossrefType "journal-article" @default.
• W3217778739 hasAuthorship W3217778739A5017777141 @default.
• W3217778739 hasAuthorship W3217778739A5057910370 @default.
• W3217778739 hasBestOaLocation W32177787392 @default.
• W3217778739 hasConcept C119857082 @default.
• W3217778739 hasConcept C120665830 @default.
• W3217778739 hasConcept C121332964 @default.
• W3217778739 hasConcept C134306372 @default.
• W3217778739 hasConcept C139807058 @default.
• W3217778739 hasConcept C153180895 @default.
• W3217778739 hasConcept C154945302 @default.
• W3217778739 hasConcept C28490314 @default.
• W3217778739 hasConcept C33923547 @default.
• W3217778739 hasConcept C36503486 @default.
• W3217778739 hasConcept C41008148 @default.
• W3217778739 hasConcept C97931131 @default.
• W3217778739 hasConceptScore W3217778739C119857082 @default.
• W3217778739 hasConceptScore W3217778739C120665830 @default.
• W3217778739 hasConceptScore W3217778739C121332964 @default.
• W3217778739 hasConceptScore W3217778739C134306372 @default.
• W3217778739 hasConceptScore W3217778739C139807058 @default.
• W3217778739 hasConceptScore W3217778739C153180895 @default.
• W3217778739 hasConceptScore W3217778739C154945302 @default.
• W3217778739 hasConceptScore W3217778739C28490314 @default.
• W3217778739 hasConceptScore W3217778739C33923547 @default.
• W3217778739 hasConceptScore W3217778739C36503486 @default.
• W3217778739 hasConceptScore W3217778739C41008148 @default.
• W3217778739 hasConceptScore W3217778739C97931131 @default.
• W3217778739 hasLocation W32177787391 @default.
• W3217778739 hasLocation W32177787392 @default.
• W3217778739 hasOpenAccess W3217778739 @default.
• W3217778739 hasPrimaryLocation W32177787391 @default.
• W3217778739 hasRelatedWork W1972656095 @default.
• W3217778739 hasRelatedWork W2024160000 @default.
• W3217778739 hasRelatedWork W2061273563 @default.
• W3217778739 hasRelatedWork W2285052147 @default.
• W3217778739 hasRelatedWork W2729514902 @default.
• W3217778739 hasRelatedWork W2743258233 @default.
• W3217778739 hasRelatedWork W2773500201 @default.
• W3217778739 hasRelatedWork W2970216048 @default.
• W3217778739 hasRelatedWork W2998168123 @default.
• W3217778739 hasRelatedWork W4287995534 @default.
• W3217778739 hasVolume "30" @default.
• W3217778739 isParatext "false" @default.
• W3217778739 isRetracted "false" @default.
• W3217778739 magId "3217778739" @default.

• next