Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313016951> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4313016951 endingPage "119289" @default.
- W4313016951 startingPage "119283" @default.
- W4313016951 abstract "The Conformer has shown impressive performance for speech enhancement by exploiting the local and global contextual information, although it requires high computational complexity and many parameters. Recently, multi-layer perceptron (MLP)-based models such as MLP-mixer and gMLP have demonstrated comparable performances with much less computational complexity in the computer vision area. These models showed that all-MLP architectures may perform as good as more advanced structures, but the nature of the MLP limits the application of these architectures to the input with a variable length such as speech and audio. In this paper, we propose the cgMLP-SE model, which is a gMLP-based architecture with convolutional token mixing modules and squeeze-and-excitation network (SENet) to utilize both local and global contextual information as in the Conformer. Specifically, the token-mixing modules in gMLP are replaced by convolutional layers, SENet-based gating is applied on top of the convolutional gating module, and additional feed-forward layers are added to make the cgMLP-SE module a macaron-like structure sandwiched by feed-forward layers like a Conformer block. Experimental results on the TIMIT-DNS noise dataset and the Voice Bank-DEMAND dataset showed that the proposed method exhibited similar speech quality and intelligibility to the Conformer with a smaller model size and less computational complexity." @default.
- W4313016951 created "2023-01-05" @default.
- W4313016951 creator A5004608249 @default.
- W4313016951 creator A5054155133 @default.
- W4313016951 creator A5073806960 @default.
- W4313016951 date "2022-01-01" @default.
- W4313016951 modified "2023-10-18" @default.
- W4313016951 title "Speech Enhancement Using MLP-Based Architecture With Convolutional Token Mixing Module and Squeeze-and-Excitation Network" @default.
- W4313016951 cites W1689711448 @default.
- W4313016951 cites W2044893557 @default.
- W4313016951 cites W2078528584 @default.
- W4313016951 cites W2094721231 @default.
- W4313016951 cites W2121973264 @default.
- W4313016951 cites W2141998673 @default.
- W4313016951 cites W2144404214 @default.
- W4313016951 cites W2603567530 @default.
- W4313016951 cites W2752782242 @default.
- W4313016951 cites W2952979007 @default.
- W4313016951 cites W3015199127 @default.
- W4313016951 cites W3017350693 @default.
- W4313016951 cites W3034421924 @default.
- W4313016951 cites W3096184950 @default.
- W4313016951 cites W3097777922 @default.
- W4313016951 cites W3162520476 @default.
- W4313016951 cites W3163838801 @default.
- W4313016951 cites W3163842642 @default.
- W4313016951 cites W3164736513 @default.
- W4313016951 cites W3173621652 @default.
- W4313016951 cites W3181056718 @default.
- W4313016951 cites W3196194966 @default.
- W4313016951 cites W3197042120 @default.
- W4313016951 cites W3197729725 @default.
- W4313016951 cites W3212754542 @default.
- W4313016951 cites W4210341744 @default.
- W4313016951 cites W4221154745 @default.
- W4313016951 cites W4224924217 @default.
- W4313016951 cites W4232282348 @default.
- W4313016951 cites W4297841897 @default.
- W4313016951 doi "https://doi.org/10.1109/access.2022.3221440" @default.
- W4313016951 hasPublicationYear "2022" @default.
- W4313016951 type Work @default.
- W4313016951 citedByCount "0" @default.
- W4313016951 crossrefType "journal-article" @default.
- W4313016951 hasAuthorship W4313016951A5004608249 @default.
- W4313016951 hasAuthorship W4313016951A5054155133 @default.
- W4313016951 hasAuthorship W4313016951A5073806960 @default.
- W4313016951 hasBestOaLocation W43130169511 @default.
- W4313016951 hasConcept C11413529 @default.
- W4313016951 hasConcept C154945302 @default.
- W4313016951 hasConcept C179717631 @default.
- W4313016951 hasConcept C179799912 @default.
- W4313016951 hasConcept C28490314 @default.
- W4313016951 hasConcept C38652104 @default.
- W4313016951 hasConcept C41008148 @default.
- W4313016951 hasConcept C48145219 @default.
- W4313016951 hasConcept C50644808 @default.
- W4313016951 hasConcept C81363708 @default.
- W4313016951 hasConceptScore W4313016951C11413529 @default.
- W4313016951 hasConceptScore W4313016951C154945302 @default.
- W4313016951 hasConceptScore W4313016951C179717631 @default.
- W4313016951 hasConceptScore W4313016951C179799912 @default.
- W4313016951 hasConceptScore W4313016951C28490314 @default.
- W4313016951 hasConceptScore W4313016951C38652104 @default.
- W4313016951 hasConceptScore W4313016951C41008148 @default.
- W4313016951 hasConceptScore W4313016951C48145219 @default.
- W4313016951 hasConceptScore W4313016951C50644808 @default.
- W4313016951 hasConceptScore W4313016951C81363708 @default.
- W4313016951 hasFunder F4320322120 @default.
- W4313016951 hasFunder F4320335489 @default.
- W4313016951 hasLocation W43130169511 @default.
- W4313016951 hasLocation W43130169512 @default.
- W4313016951 hasOpenAccess W4313016951 @default.
- W4313016951 hasPrimaryLocation W43130169511 @default.
- W4313016951 hasRelatedWork W1985412924 @default.
- W4313016951 hasRelatedWork W2375389409 @default.
- W4313016951 hasRelatedWork W2488051804 @default.
- W4313016951 hasRelatedWork W2735477435 @default.
- W4313016951 hasRelatedWork W2807436399 @default.
- W4313016951 hasRelatedWork W3016958897 @default.
- W4313016951 hasRelatedWork W3045739591 @default.
- W4313016951 hasRelatedWork W3181746755 @default.
- W4313016951 hasRelatedWork W4283379348 @default.
- W4313016951 hasRelatedWork W4312417841 @default.
- W4313016951 hasVolume "10" @default.
- W4313016951 isParatext "false" @default.
- W4313016951 isRetracted "false" @default.
- W4313016951 workType "article" @default.