Matches in SemOpenAlex for { <https://semopenalex.org/work/W3049430014> ?p ?o ?g. }
Showing items 1 to 97 of
97
with 100 items per page.
- W3049430014 endingPage "2499" @default.
- W3049430014 startingPage "2489" @default.
- W3049430014 abstract "In recent years, supervised approaches using deep neural networks (DNNs) have become the mainstream for speech enhancement. It has been established that DNNs generalize well to untrained noises and speakers if trained using a large number of noises and speakers. However, we find that DNNs fail to generalize to new speech corpora in low signal-to-noise ratio (SNR) conditions. In this work, we establish that the lack of generalization is mainly due to the channel mismatch, i.e. different recording conditions between the trained and untrained corpus. Additionally, we observe that traditional channel normalization techniques are not effective in improving cross-corpus generalization. Further, we evaluate publicly available datasets that are promising for generalization. We find one particular corpus to be significantly better than others. Finally, we find that using a smaller frame shift in short-time processing of speech can significantly improve cross-corpus generalization. The proposed techniques to address cross-corpus generalization include channel normalization, better training corpus, and smaller frame shift in short-time Fourier transform (STFT). These techniques together improve the objective intelligibility and quality scores on untrained corpora significantly." @default.
- W3049430014 created "2020-08-21" @default.
- W3049430014 creator A5050492887 @default.
- W3049430014 creator A5051837453 @default.
- W3049430014 date "2020-01-01" @default.
- W3049430014 modified "2023-10-03" @default.
- W3049430014 title "On Cross-Corpus Generalization of Deep Learning Based Speech Enhancement" @default.
- W3049430014 cites W1494198834 @default.
- W3049430014 cites W1552314771 @default.
- W3049430014 cites W1974387177 @default.
- W3049430014 cites W2013608223 @default.
- W3049430014 cites W2024490156 @default.
- W3049430014 cites W2029757964 @default.
- W3049430014 cites W2044893557 @default.
- W3049430014 cites W2069681747 @default.
- W3049430014 cites W2069976350 @default.
- W3049430014 cites W2112180173 @default.
- W3049430014 cites W2114218864 @default.
- W3049430014 cites W2128653836 @default.
- W3049430014 cites W2137075158 @default.
- W3049430014 cites W2141998673 @default.
- W3049430014 cites W2153894152 @default.
- W3049430014 cites W2168379380 @default.
- W3049430014 cites W2168548156 @default.
- W3049430014 cites W2168961642 @default.
- W3049430014 cites W2221409856 @default.
- W3049430014 cites W2291877678 @default.
- W3049430014 cites W2364134690 @default.
- W3049430014 cites W2678916739 @default.
- W3049430014 cites W2808631503 @default.
- W3049430014 cites W2890820256 @default.
- W3049430014 cites W2892356933 @default.
- W3049430014 cites W2897371647 @default.
- W3049430014 cites W2937484199 @default.
- W3049430014 cites W2938901746 @default.
- W3049430014 cites W2940275453 @default.
- W3049430014 cites W2943554574 @default.
- W3049430014 cites W2962866211 @default.
- W3049430014 cites W2963103134 @default.
- W3049430014 cites W2963828919 @default.
- W3049430014 cites W2991361823 @default.
- W3049430014 cites W4231807801 @default.
- W3049430014 cites W4253928870 @default.
- W3049430014 doi "https://doi.org/10.1109/taslp.2020.3016487" @default.
- W3049430014 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7971413" @default.
- W3049430014 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33748327" @default.
- W3049430014 hasPublicationYear "2020" @default.
- W3049430014 type Work @default.
- W3049430014 sameAs 3049430014 @default.
- W3049430014 citedByCount "30" @default.
- W3049430014 countsByYear W30494300142020 @default.
- W3049430014 countsByYear W30494300142021 @default.
- W3049430014 countsByYear W30494300142022 @default.
- W3049430014 countsByYear W30494300142023 @default.
- W3049430014 crossrefType "journal-article" @default.
- W3049430014 hasAuthorship W3049430014A5050492887 @default.
- W3049430014 hasAuthorship W3049430014A5051837453 @default.
- W3049430014 hasBestOaLocation W30494300142 @default.
- W3049430014 hasConcept C108583219 @default.
- W3049430014 hasConcept C134306372 @default.
- W3049430014 hasConcept C154945302 @default.
- W3049430014 hasConcept C177148314 @default.
- W3049430014 hasConcept C204321447 @default.
- W3049430014 hasConcept C28490314 @default.
- W3049430014 hasConcept C33923547 @default.
- W3049430014 hasConcept C41008148 @default.
- W3049430014 hasConceptScore W3049430014C108583219 @default.
- W3049430014 hasConceptScore W3049430014C134306372 @default.
- W3049430014 hasConceptScore W3049430014C154945302 @default.
- W3049430014 hasConceptScore W3049430014C177148314 @default.
- W3049430014 hasConceptScore W3049430014C204321447 @default.
- W3049430014 hasConceptScore W3049430014C28490314 @default.
- W3049430014 hasConceptScore W3049430014C33923547 @default.
- W3049430014 hasConceptScore W3049430014C41008148 @default.
- W3049430014 hasFunder F4320337352 @default.
- W3049430014 hasLocation W30494300141 @default.
- W3049430014 hasLocation W30494300142 @default.
- W3049430014 hasLocation W30494300143 @default.
- W3049430014 hasOpenAccess W3049430014 @default.
- W3049430014 hasPrimaryLocation W30494300141 @default.
- W3049430014 hasRelatedWork W2126887587 @default.
- W3049430014 hasRelatedWork W2731899572 @default.
- W3049430014 hasRelatedWork W2939353110 @default.
- W3049430014 hasRelatedWork W2948658236 @default.
- W3049430014 hasRelatedWork W2993873509 @default.
- W3049430014 hasRelatedWork W3009238340 @default.
- W3049430014 hasRelatedWork W3020224787 @default.
- W3049430014 hasRelatedWork W3215138031 @default.
- W3049430014 hasRelatedWork W4220740160 @default.
- W3049430014 hasRelatedWork W4230611425 @default.
- W3049430014 hasVolume "28" @default.
- W3049430014 isParatext "false" @default.
- W3049430014 isRetracted "false" @default.
- W3049430014 magId "3049430014" @default.
- W3049430014 workType "article" @default.