Matches in SemOpenAlex for { <https://semopenalex.org/work/W3138320152> ?p ?o ?g. }
Showing items 1 to 94 of
94
with 100 items per page.
- W3138320152 abstract "In this paper, we propose a transformer-based architecture, called two-stage transformer neural network (TSTNN) for end-to-end speech denoising in the time domain. The proposed model is composed of an encoder, a two-stage transformer module (TSTM), a masking module and a decoder. The encoder maps input noisy speech into feature representation. The TSTM exploits four stacked two-stage transformer blocks to efficiently extract local and global information from the encoder output stage by stage. The masking module creates a mask which will be multiplied with the encoder output. Finally, the decoder uses the masked encoder feature to reconstruct the enhanced speech. Experimental results on the benchmark dataset show that the TSTNN outperforms most state-of-the-art models in time or frequency domain while having significantly lower model complexity." @default.
- W3138320152 created "2021-03-29" @default.
- W3138320152 creator A5033247734 @default.
- W3138320152 creator A5041148892 @default.
- W3138320152 creator A5063950661 @default.
- W3138320152 date "2021-03-17" @default.
- W3138320152 modified "2023-10-16" @default.
- W3138320152 title "TSTNN: Two-stage Transformer based Neural Network for Speech Enhancement in the Time Domain" @default.
- W3138320152 cites W1495679096 @default.
- W3138320152 cites W1677182931 @default.
- W3138320152 cites W1983108229 @default.
- W3138320152 cites W2094721231 @default.
- W3138320152 cites W2141998673 @default.
- W3138320152 cites W2144404214 @default.
- W3138320152 cites W2403555780 @default.
- W3138320152 cites W2476548250 @default.
- W3138320152 cites W2600556233 @default.
- W3138320152 cites W2603567530 @default.
- W3138320152 cites W2774389566 @default.
- W3138320152 cites W2802304149 @default.
- W3138320152 cites W2889134433 @default.
- W3138320152 cites W2902132730 @default.
- W3138320152 cites W2906042495 @default.
- W3138320152 cites W2921144622 @default.
- W3138320152 cites W2937484199 @default.
- W3138320152 cites W2943895317 @default.
- W3138320152 cites W2963103134 @default.
- W3138320152 cites W2963403868 @default.
- W3138320152 cites W2963840672 @default.
- W3138320152 cites W2964089206 @default.
- W3138320152 cites W3015197852 @default.
- W3138320152 cites W3015199127 @default.
- W3138320152 cites W3016129867 @default.
- W3138320152 cites W3036367741 @default.
- W3138320152 cites W3045904949 @default.
- W3138320152 doi "https://doi.org/10.48550/arxiv.2103.09963" @default.
- W3138320152 hasPublicationYear "2021" @default.
- W3138320152 type Work @default.
- W3138320152 sameAs 3138320152 @default.
- W3138320152 citedByCount "0" @default.
- W3138320152 crossrefType "posted-content" @default.
- W3138320152 hasAuthorship W3138320152A5033247734 @default.
- W3138320152 hasAuthorship W3138320152A5041148892 @default.
- W3138320152 hasAuthorship W3138320152A5063950661 @default.
- W3138320152 hasBestOaLocation W31383201521 @default.
- W3138320152 hasConcept C103824480 @default.
- W3138320152 hasConcept C111919701 @default.
- W3138320152 hasConcept C118505674 @default.
- W3138320152 hasConcept C119599485 @default.
- W3138320152 hasConcept C127413603 @default.
- W3138320152 hasConcept C13280743 @default.
- W3138320152 hasConcept C153180895 @default.
- W3138320152 hasConcept C154945302 @default.
- W3138320152 hasConcept C165801399 @default.
- W3138320152 hasConcept C185798385 @default.
- W3138320152 hasConcept C205649164 @default.
- W3138320152 hasConcept C28490314 @default.
- W3138320152 hasConcept C31972630 @default.
- W3138320152 hasConcept C41008148 @default.
- W3138320152 hasConcept C50644808 @default.
- W3138320152 hasConcept C66322947 @default.
- W3138320152 hasConceptScore W3138320152C103824480 @default.
- W3138320152 hasConceptScore W3138320152C111919701 @default.
- W3138320152 hasConceptScore W3138320152C118505674 @default.
- W3138320152 hasConceptScore W3138320152C119599485 @default.
- W3138320152 hasConceptScore W3138320152C127413603 @default.
- W3138320152 hasConceptScore W3138320152C13280743 @default.
- W3138320152 hasConceptScore W3138320152C153180895 @default.
- W3138320152 hasConceptScore W3138320152C154945302 @default.
- W3138320152 hasConceptScore W3138320152C165801399 @default.
- W3138320152 hasConceptScore W3138320152C185798385 @default.
- W3138320152 hasConceptScore W3138320152C205649164 @default.
- W3138320152 hasConceptScore W3138320152C28490314 @default.
- W3138320152 hasConceptScore W3138320152C31972630 @default.
- W3138320152 hasConceptScore W3138320152C41008148 @default.
- W3138320152 hasConceptScore W3138320152C50644808 @default.
- W3138320152 hasConceptScore W3138320152C66322947 @default.
- W3138320152 hasLocation W31383201521 @default.
- W3138320152 hasOpenAccess W3138320152 @default.
- W3138320152 hasPrimaryLocation W31383201521 @default.
- W3138320152 hasRelatedWork W11765363 @default.
- W3138320152 hasRelatedWork W2151333624 @default.
- W3138320152 hasRelatedWork W2526733715 @default.
- W3138320152 hasRelatedWork W2892009249 @default.
- W3138320152 hasRelatedWork W2963453742 @default.
- W3138320152 hasRelatedWork W3036598372 @default.
- W3138320152 hasRelatedWork W3047941943 @default.
- W3138320152 hasRelatedWork W3157428800 @default.
- W3138320152 hasRelatedWork W3196667351 @default.
- W3138320152 hasRelatedWork W4287186213 @default.
- W3138320152 isParatext "false" @default.
- W3138320152 isRetracted "false" @default.
- W3138320152 magId "3138320152" @default.
- W3138320152 workType "article" @default.