Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891607145> ?p ?o ?g. }
Showing items 1 to 85 of
85
with 100 items per page.
- W2891607145 abstract "Human listeners often have difficulties understanding speech in the presence of background noise in the real world. Recently, supervised learning based speech enhancement approaches have achieved substantial success, and show significant improvements over the conventional approaches. However, existing supervised learning based approaches often try to minimize the mean squared error between the enhanced output and the pre-defined training target (e.g., the log power spectrum of clean speech), even though the purpose of such speech enhancement is to improve speech understanding in noise. In this paper, we propose a new deep neural networks based enhancement approach by incorporating a speech perception model into the loss function. Specifically, we use the short-time objective intelligibility metric in the loss in addition to the mean squared error. Optimizing the proposed perceptually guided loss is expected to improve speech intelligibility further. Systematic evaluations show that our proposed approach is able to improve speech intelligibility in a wide range of signal-to-noise ratios and noise types while maintaining speech quality." @default.
- W2891607145 created "2018-09-27" @default.
- W2891607145 creator A5035701638 @default.
- W2891607145 creator A5040562171 @default.
- W2891607145 creator A5076372893 @default.
- W2891607145 creator A5091775011 @default.
- W2891607145 date "2018-04-01" @default.
- W2891607145 modified "2023-10-17" @default.
- W2891607145 title "Perceptually Guided Speech Enhancement Using Deep Neural Networks" @default.
- W2891607145 cites W1482149378 @default.
- W2891607145 cites W1516630152 @default.
- W2891607145 cites W1552314771 @default.
- W2891607145 cites W1974387177 @default.
- W2891607145 cites W1992475611 @default.
- W2891607145 cites W2044893557 @default.
- W2891607145 cites W2046869671 @default.
- W2891607145 cites W2069681747 @default.
- W2891607145 cites W2105921478 @default.
- W2891607145 cites W2121061347 @default.
- W2891607145 cites W2127851351 @default.
- W2891607145 cites W2141998673 @default.
- W2891607145 cites W2291877678 @default.
- W2891607145 cites W4231807801 @default.
- W2891607145 doi "https://doi.org/10.1109/icassp.2018.8462593" @default.
- W2891607145 hasPublicationYear "2018" @default.
- W2891607145 type Work @default.
- W2891607145 sameAs 2891607145 @default.
- W2891607145 citedByCount "54" @default.
- W2891607145 countsByYear W28916071452019 @default.
- W2891607145 countsByYear W28916071452020 @default.
- W2891607145 countsByYear W28916071452021 @default.
- W2891607145 countsByYear W28916071452022 @default.
- W2891607145 countsByYear W28916071452023 @default.
- W2891607145 crossrefType "proceedings-article" @default.
- W2891607145 hasAuthorship W2891607145A5035701638 @default.
- W2891607145 hasAuthorship W2891607145A5040562171 @default.
- W2891607145 hasAuthorship W2891607145A5076372893 @default.
- W2891607145 hasAuthorship W2891607145A5091775011 @default.
- W2891607145 hasConcept C105795698 @default.
- W2891607145 hasConcept C108583219 @default.
- W2891607145 hasConcept C111472728 @default.
- W2891607145 hasConcept C115961682 @default.
- W2891607145 hasConcept C138885662 @default.
- W2891607145 hasConcept C139945424 @default.
- W2891607145 hasConcept C154945302 @default.
- W2891607145 hasConcept C163294075 @default.
- W2891607145 hasConcept C2776182073 @default.
- W2891607145 hasConcept C28490314 @default.
- W2891607145 hasConcept C33923547 @default.
- W2891607145 hasConcept C41008148 @default.
- W2891607145 hasConcept C50644808 @default.
- W2891607145 hasConcept C60048801 @default.
- W2891607145 hasConcept C99498987 @default.
- W2891607145 hasConceptScore W2891607145C105795698 @default.
- W2891607145 hasConceptScore W2891607145C108583219 @default.
- W2891607145 hasConceptScore W2891607145C111472728 @default.
- W2891607145 hasConceptScore W2891607145C115961682 @default.
- W2891607145 hasConceptScore W2891607145C138885662 @default.
- W2891607145 hasConceptScore W2891607145C139945424 @default.
- W2891607145 hasConceptScore W2891607145C154945302 @default.
- W2891607145 hasConceptScore W2891607145C163294075 @default.
- W2891607145 hasConceptScore W2891607145C2776182073 @default.
- W2891607145 hasConceptScore W2891607145C28490314 @default.
- W2891607145 hasConceptScore W2891607145C33923547 @default.
- W2891607145 hasConceptScore W2891607145C41008148 @default.
- W2891607145 hasConceptScore W2891607145C50644808 @default.
- W2891607145 hasConceptScore W2891607145C60048801 @default.
- W2891607145 hasConceptScore W2891607145C99498987 @default.
- W2891607145 hasLocation W28916071451 @default.
- W2891607145 hasOpenAccess W2891607145 @default.
- W2891607145 hasPrimaryLocation W28916071451 @default.
- W2891607145 hasRelatedWork W1505346162 @default.
- W2891607145 hasRelatedWork W1600259599 @default.
- W2891607145 hasRelatedWork W1986772939 @default.
- W2891607145 hasRelatedWork W2037635165 @default.
- W2891607145 hasRelatedWork W2147952370 @default.
- W2891607145 hasRelatedWork W2187949121 @default.
- W2891607145 hasRelatedWork W2891607145 @default.
- W2891607145 hasRelatedWork W3129072390 @default.
- W2891607145 hasRelatedWork W4200562864 @default.
- W2891607145 hasRelatedWork W4221152531 @default.
- W2891607145 isParatext "false" @default.
- W2891607145 isRetracted "false" @default.
- W2891607145 magId "2891607145" @default.
- W2891607145 workType "article" @default.