Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897725921> ?p ?o ?g. }
Showing items 1 to 87 of
87
with 100 items per page.
- W2897725921 abstract "In this paper, we compare different deep neural networks (DNN) in extracting speech signals from competing speakers in room environments, including the conventional fully-connected multilayer perception (MLP) network, convolutional neural network (CNN), recurrent neural network (RNN), and the recently proposed capsule network (CapsNet). Each DNN takes input of both spectral features and converted spatial features that are robust to position mismatch, and outputs the separation mask for target source estimation. In addition, a psychacoustically-motivated objective function is integrated in each DNN, which explores perceptual importance of each TF unit in the training process. Objective evaluations are performed on the separated sounds using the converged models, in terms of PESQ, SDR as well as STOI. Overall, all the implemented DNNs have greatly improved the quality and speech intelligibility of the embedded target source as compared to the original recordings. In particular, bidirectional RNN, either along the temporal direction or along the frequency bins, outperforms the other DNN structures with consistent improvement." @default.
- W2897725921 created "2018-10-26" @default.
- W2897725921 creator A5009941201 @default.
- W2897725921 creator A5037691180 @default.
- W2897725921 creator A5074138677 @default.
- W2897725921 creator A5086899221 @default.
- W2897725921 date "2018-10-01" @default.
- W2897725921 modified "2023-09-23" @default.
- W2897725921 title "A Performance Evaluation of Several Deep Neural Networks for Reverberant Speech Separation" @default.
- W2897725921 cites W1552314771 @default.
- W2897725921 cites W1836465849 @default.
- W2897725921 cites W1897240248 @default.
- W2897725921 cites W1992879732 @default.
- W2897725921 cites W2031647436 @default.
- W2897725921 cites W2046317813 @default.
- W2897725921 cites W2067295501 @default.
- W2897725921 cites W2221409856 @default.
- W2897725921 cites W2537586743 @default.
- W2897725921 cites W2587994092 @default.
- W2897725921 cites W2599685671 @default.
- W2897725921 cites W2605138598 @default.
- W2897725921 cites W2620812332 @default.
- W2897725921 cites W2797579321 @default.
- W2897725921 cites W2962866211 @default.
- W2897725921 cites W2963703618 @default.
- W2897725921 cites W2964121744 @default.
- W2897725921 cites W3023071679 @default.
- W2897725921 doi "https://doi.org/10.1109/acssc.2018.8645219" @default.
- W2897725921 hasPublicationYear "2018" @default.
- W2897725921 type Work @default.
- W2897725921 sameAs 2897725921 @default.
- W2897725921 citedByCount "1" @default.
- W2897725921 countsByYear W28977259212020 @default.
- W2897725921 crossrefType "proceedings-article" @default.
- W2897725921 hasAuthorship W2897725921A5009941201 @default.
- W2897725921 hasAuthorship W2897725921A5037691180 @default.
- W2897725921 hasAuthorship W2897725921A5074138677 @default.
- W2897725921 hasAuthorship W2897725921A5086899221 @default.
- W2897725921 hasBestOaLocation W28977259212 @default.
- W2897725921 hasConcept C103734657 @default.
- W2897725921 hasConcept C108583219 @default.
- W2897725921 hasConcept C111472728 @default.
- W2897725921 hasConcept C138885662 @default.
- W2897725921 hasConcept C147168706 @default.
- W2897725921 hasConcept C153180895 @default.
- W2897725921 hasConcept C154945302 @default.
- W2897725921 hasConcept C163294075 @default.
- W2897725921 hasConcept C175202392 @default.
- W2897725921 hasConcept C2776182073 @default.
- W2897725921 hasConcept C28490314 @default.
- W2897725921 hasConcept C41008148 @default.
- W2897725921 hasConcept C50644808 @default.
- W2897725921 hasConcept C60048801 @default.
- W2897725921 hasConcept C81363708 @default.
- W2897725921 hasConceptScore W2897725921C103734657 @default.
- W2897725921 hasConceptScore W2897725921C108583219 @default.
- W2897725921 hasConceptScore W2897725921C111472728 @default.
- W2897725921 hasConceptScore W2897725921C138885662 @default.
- W2897725921 hasConceptScore W2897725921C147168706 @default.
- W2897725921 hasConceptScore W2897725921C153180895 @default.
- W2897725921 hasConceptScore W2897725921C154945302 @default.
- W2897725921 hasConceptScore W2897725921C163294075 @default.
- W2897725921 hasConceptScore W2897725921C175202392 @default.
- W2897725921 hasConceptScore W2897725921C2776182073 @default.
- W2897725921 hasConceptScore W2897725921C28490314 @default.
- W2897725921 hasConceptScore W2897725921C41008148 @default.
- W2897725921 hasConceptScore W2897725921C50644808 @default.
- W2897725921 hasConceptScore W2897725921C60048801 @default.
- W2897725921 hasConceptScore W2897725921C81363708 @default.
- W2897725921 hasLocation W28977259211 @default.
- W2897725921 hasLocation W28977259212 @default.
- W2897725921 hasOpenAccess W2897725921 @default.
- W2897725921 hasPrimaryLocation W28977259211 @default.
- W2897725921 hasRelatedWork W2037635165 @default.
- W2897725921 hasRelatedWork W2140410589 @default.
- W2897725921 hasRelatedWork W2787650696 @default.
- W2897725921 hasRelatedWork W2897725921 @default.
- W2897725921 hasRelatedWork W2962892438 @default.
- W2897725921 hasRelatedWork W3168779539 @default.
- W2897725921 hasRelatedWork W3175173795 @default.
- W2897725921 hasRelatedWork W3209446892 @default.
- W2897725921 hasRelatedWork W4200562864 @default.
- W2897725921 hasRelatedWork W4244026860 @default.
- W2897725921 isParatext "false" @default.
- W2897725921 isRetracted "false" @default.
- W2897725921 magId "2897725921" @default.
- W2897725921 workType "article" @default.