FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3015969340> ?p ?o ?g. }

• W3015969340 endingPage "1949" @default.
• W3015969340 startingPage "1937" @default.
• W3015969340 abstract "In this paper, we address the generalization of deep neural network (DNN) based speech enhancement to unseen noise conditions for the case that training data is limited in size and diversity. To gain more insights, we analyze the generalization with respect to (1) the size and diversity of the training data, (2) different network architectures, and (3) the chosen features. To address (1), we train networks on the Hu noise corpus (limited size), the CHiME 3 noise corpus (limited diversity) and also propose a large and diverse dataset collected based on freely available sounds. To address (2), we compare a fully-connected feed-forward and a long short-term memory (LSTM) architecture. To address (3), we compare three input features, namely logarithmized noisy periodograms, noise aware training (NAT) and the proposed signal-to-noise ratio based noise aware training (SNR-NAT). We confirm that rich training data and improved network architectures help DNNs to generalize. Furthermore, we show via experimental results and an analysis using t-distributed stochastic neighbor embedding (t-SNE) that the proposed SNR-NAT features yield robust and level independent results in unseen noise even with simple network architectures and when trained on only small datasets, which is the key contribution of this paper." @default.
• W3015969340 created "2020-04-17" @default.
• W3015969340 creator A5041108140 @default.
• W3015969340 creator A5087022569 @default.
• W3015969340 date "2021-01-01" @default.
• W3015969340 modified "2023-09-30" @default.
• W3015969340 title "SNR-Based Features and Diverse Training Data for Robust DNN-Based Speech Enhancement" @default.
• W3015969340 cites W1482149378 @default.
• W3015969340 cites W1897240248 @default.
• W3015969340 cites W1963950237 @default.
• W3015969340 cites W1964509438 @default.
• W3015969340 cites W1974387177 @default.
• W3015969340 cites W1988115241 @default.
• W3015969340 cites W1989364685 @default.
• W3015969340 cites W1995190897 @default.
• W3015969340 cites W2013608223 @default.
• W3015969340 cites W2016254085 @default.
• W3015969340 cites W2016874282 @default.
• W3015969340 cites W2018026382 @default.
• W3015969340 cites W2025515167 @default.
• W3015969340 cites W2025611902 @default.
• W3015969340 cites W2042071805 @default.
• W3015969340 cites W2044893557 @default.
• W3015969340 cites W2046869671 @default.
• W3015969340 cites W2051428568 @default.
• W3015969340 cites W2062164080 @default.
• W3015969340 cites W2063157319 @default.
• W3015969340 cites W2064675550 @default.
• W3015969340 cites W2069486713 @default.
• W3015969340 cites W2069681747 @default.
• W3015969340 cites W2103135337 @default.
• W3015969340 cites W2103747981 @default.
• W3015969340 cites W2114780615 @default.
• W3015969340 cites W2127180234 @default.
• W3015969340 cites W2128653836 @default.
• W3015969340 cites W2130476329 @default.
• W3015969340 cites W2144739192 @default.
• W3015969340 cites W2145310550 @default.
• W3015969340 cites W2148154194 @default.
• W3015969340 cites W2149535104 @default.
• W3015969340 cites W2158336491 @default.
• W3015969340 cites W2163195137 @default.
• W3015969340 cites W2164432048 @default.
• W3015969340 cites W2164715564 @default.
• W3015969340 cites W2167090878 @default.
• W3015969340 cites W2169147844 @default.
• W3015969340 cites W2236045036 @default.
• W3015969340 cites W2285420822 @default.
• W3015969340 cites W2289394825 @default.
• W3015969340 cites W2291877678 @default.
• W3015969340 cites W2364134690 @default.
• W3015969340 cites W2507424718 @default.
• W3015969340 cites W2516001803 @default.
• W3015969340 cites W2532854561 @default.
• W3015969340 cites W2550397165 @default.
• W3015969340 cites W2559809918 @default.
• W3015969340 cites W2560174002 @default.
• W3015969340 cites W2605589342 @default.
• W3015969340 cites W2606806821 @default.
• W3015969340 cites W2678916739 @default.
• W3015969340 cites W2746457594 @default.
• W3015969340 cites W2747161606 @default.
• W3015969340 cites W2791874741 @default.
• W3015969340 cites W2938737578 @default.
• W3015969340 cites W2962843322 @default.
• W3015969340 cites W2962946126 @default.
• W3015969340 cites W2963341071 @default.
• W3015969340 cites W2963453742 @default.
• W3015969340 cites W3049430014 @default.
• W3015969340 cites W3147539069 @default.
• W3015969340 cites W4256217162 @default.
• W3015969340 doi "https://doi.org/10.1109/taslp.2021.3082702" @default.
• W3015969340 hasPublicationYear "2021" @default.
• W3015969340 type Work @default.
• W3015969340 sameAs 3015969340 @default.
• W3015969340 citedByCount "10" @default.
• W3015969340 countsByYear W30159693402021 @default.
• W3015969340 countsByYear W30159693402022 @default.
• W3015969340 countsByYear W30159693402023 @default.
• W3015969340 crossrefType "journal-article" @default.
• W3015969340 hasAuthorship W3015969340A5041108140 @default.
• W3015969340 hasAuthorship W3015969340A5087022569 @default.
• W3015969340 hasBestOaLocation W30159693402 @default.
• W3015969340 hasConcept C115961682 @default.
• W3015969340 hasConcept C119857082 @default.
• W3015969340 hasConcept C121332964 @default.
• W3015969340 hasConcept C134306372 @default.
• W3015969340 hasConcept C153180895 @default.
• W3015969340 hasConcept C153294291 @default.
• W3015969340 hasConcept C154945302 @default.
• W3015969340 hasConcept C177148314 @default.
• W3015969340 hasConcept C193415008 @default.
• W3015969340 hasConcept C26517878 @default.
• W3015969340 hasConcept C2777211547 @default.
• W3015969340 hasConcept C28490314 @default.
• W3015969340 hasConcept C2984842247 @default.
• W3015969340 hasConcept C33923547 @default.
• W3015969340 hasConcept C38652104 @default.

• next