FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3167256084> ?p ?o ?g. }

• W3167256084 endingPage "84970" @default.
• W3167256084 startingPage "84956" @default.
• W3167256084 abstract "Localization in reverberant environments remains an open challenge. Recently, supervised learning approaches have demonstrated very promising results in addressing reverberation. However, even with large data volumes, the number of labels available for supervised learning in such environments is usually small. We propose to address this issue with a semi-supervised learning (SSL) approach, based on deep generative modeling. Our chosen deep generative model, the variational autoencoder (VAE), is trained to generate the phase of relative transfer functions (RTFs) between microphones. In parallel, a direction of arrival (DOA) classifier network based on RTF-phase is also trained. The joint generative and discriminative model, deemed VAE-SSL, is trained using labeled and unlabeled RTF-phase sequences. In learning to generate and classify the sequences, the VAE-SSL extracts the physical causes of the RTF-phase (i.e., source location) from distracting signal characteristics such as noise and speech activity. This facilitates effective end-to-end operation of the VAE-SSL, which requires minimal preprocessing of RTF-phase. VAE-SSL is compared with two signal processing-based approaches, steered response power with phase transform (SRP-PHAT) and MUltiple SIgnal Classification (MUSIC), as well as fully supervised CNNs. The approaches are compared using data from two real acoustic environments - one of which was recently obtained at Technical University of Denmark specifically for our study. We find that VAE-SSL can outperform the conventional approaches and the CNN in label-limited scenarios. Further, the trained VAE-SSL system can generate new RTF-phase samples which capture the physics of the acoustic environment. Thus, the generative modeling in VAE-SSL provides a means of interpreting the learned representations. To the best of our knowledge, this paper presents the first approach to modeling the physics of acoustic propagation using deep generative modeling." @default.
• W3167256084 created "2021-06-22" @default.
• W3167256084 creator A5001052072 @default.
• W3167256084 creator A5004718675 @default.
• W3167256084 creator A5018064114 @default.
• W3167256084 creator A5048037495 @default.
• W3167256084 date "2021-01-01" @default.
• W3167256084 modified "2023-09-27" @default.
• W3167256084 title "Semi-Supervised Source Localization in Reverberant Environments With Deep Generative Modeling" @default.
• W3167256084 cites W1485388545 @default.
• W3167256084 cites W1494198834 @default.
• W3167256084 cites W1555217905 @default.
• W3167256084 cites W1772191655 @default.
• W3167256084 cites W1981284917 @default.
• W3167256084 cites W1996304098 @default.
• W3167256084 cites W2113638573 @default.
• W3167256084 cites W2114863372 @default.
• W3167256084 cites W2130357996 @default.
• W3167256084 cites W2130994896 @default.
• W3167256084 cites W2147665979 @default.
• W3167256084 cites W2163922914 @default.
• W3167256084 cites W2555915854 @default.
• W3167256084 cites W2568308529 @default.
• W3167256084 cites W2763188033 @default.
• W3167256084 cites W2810934215 @default.
• W3167256084 cites W2885219692 @default.
• W3167256084 cites W2901552243 @default.
• W3167256084 cites W2902304208 @default.
• W3167256084 cites W2918296821 @default.
• W3167256084 cites W2931364255 @default.
• W3167256084 cites W2939592795 @default.
• W3167256084 cites W2948978827 @default.
• W3167256084 cites W2963321191 @default.
• W3167256084 cites W2997989584 @default.
• W3167256084 cites W3014801121 @default.
• W3167256084 cites W3031005739 @default.
• W3167256084 cites W3034664488 @default.
• W3167256084 cites W3092632462 @default.
• W3167256084 cites W3101380508 @default.
• W3167256084 cites W3102937397 @default.
• W3167256084 cites W3105475478 @default.
• W3167256084 cites W3109222830 @default.
• W3167256084 cites W3129910207 @default.
• W3167256084 doi "https://doi.org/10.1109/access.2021.3087697" @default.
• W3167256084 hasPublicationYear "2021" @default.
• W3167256084 type Work @default.
• W3167256084 sameAs 3167256084 @default.
• W3167256084 citedByCount "19" @default.
• W3167256084 countsByYear W31672560842021 @default.
• W3167256084 countsByYear W31672560842022 @default.
• W3167256084 countsByYear W31672560842023 @default.
• W3167256084 crossrefType "journal-article" @default.
• W3167256084 hasAuthorship W3167256084A5001052072 @default.
• W3167256084 hasAuthorship W3167256084A5004718675 @default.
• W3167256084 hasAuthorship W3167256084A5018064114 @default.
• W3167256084 hasAuthorship W3167256084A5048037495 @default.
• W3167256084 hasBestOaLocation W31672560841 @default.
• W3167256084 hasConcept C101738243 @default.
• W3167256084 hasConcept C108583219 @default.
• W3167256084 hasConcept C119857082 @default.
• W3167256084 hasConcept C121332964 @default.
• W3167256084 hasConcept C136389625 @default.
• W3167256084 hasConcept C150899416 @default.
• W3167256084 hasConcept C153180895 @default.
• W3167256084 hasConcept C154945302 @default.
• W3167256084 hasConcept C167966045 @default.
• W3167256084 hasConcept C24890656 @default.
• W3167256084 hasConcept C28490314 @default.
• W3167256084 hasConcept C34736171 @default.
• W3167256084 hasConcept C39890363 @default.
• W3167256084 hasConcept C41008148 @default.
• W3167256084 hasConcept C50644808 @default.
• W3167256084 hasConcept C95623464 @default.
• W3167256084 hasConcept C95851461 @default.
• W3167256084 hasConcept C97931131 @default.
• W3167256084 hasConceptScore W3167256084C101738243 @default.
• W3167256084 hasConceptScore W3167256084C108583219 @default.
• W3167256084 hasConceptScore W3167256084C119857082 @default.
• W3167256084 hasConceptScore W3167256084C121332964 @default.
• W3167256084 hasConceptScore W3167256084C136389625 @default.
• W3167256084 hasConceptScore W3167256084C150899416 @default.
• W3167256084 hasConceptScore W3167256084C153180895 @default.
• W3167256084 hasConceptScore W3167256084C154945302 @default.
• W3167256084 hasConceptScore W3167256084C167966045 @default.
• W3167256084 hasConceptScore W3167256084C24890656 @default.
• W3167256084 hasConceptScore W3167256084C28490314 @default.
• W3167256084 hasConceptScore W3167256084C34736171 @default.
• W3167256084 hasConceptScore W3167256084C39890363 @default.
• W3167256084 hasConceptScore W3167256084C41008148 @default.
• W3167256084 hasConceptScore W3167256084C50644808 @default.
• W3167256084 hasConceptScore W3167256084C95623464 @default.
• W3167256084 hasConceptScore W3167256084C95851461 @default.
• W3167256084 hasConceptScore W3167256084C97931131 @default.
• W3167256084 hasFunder F4320337345 @default.
• W3167256084 hasLocation W31672560841 @default.
• W3167256084 hasLocation W31672560842 @default.
• W3167256084 hasLocation W31672560843 @default.
• W3167256084 hasLocation W31672560844 @default.

• next