Matches in SemOpenAlex for { <https://semopenalex.org/work/W4214967232> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4214967232 abstract "The front-end module in multi-channel automatic speech recognition (ASR) systems mainly use microphone array techniques to produce enhanced signals in noisy conditions with reverberation and echos. Recently, neural network (NN) based front-end has shown promising improvement over the conventional signal processing methods. In this paper, we propose to adopt the architecture of deep complex Unet (DCUnet) - a powerful complex-valued Unet-structured speech enhancement model - as the front-end of the multi-channel acoustic model, and integrate them in a multi-task learning (MTL) framework along with cascaded framework for comparison. Meanwhile, we investigate the proposed methods with several training strategies to improve the recognition accuracy on the 1000-hours real-world XiaoMi smart speaker data with echos. Experiments show that our proposed DCUnet-MTL method brings about 12.2% relative character error rate (CER) reduction compared with the traditional approach with array processing plus single-channel acoustic model. It also achieves superior performance than the recently proposed neural beamforming method." @default.
- W4214967232 created "2022-03-05" @default.
- W4214967232 creator A5013004114 @default.
- W4214967232 creator A5028067869 @default.
- W4214967232 creator A5029110813 @default.
- W4214967232 creator A5033976028 @default.
- W4214967232 creator A5065721838 @default.
- W4214967232 creator A5077264905 @default.
- W4214967232 creator A5087994947 @default.
- W4214967232 date "2020-11-17" @default.
- W4214967232 modified "2023-10-16" @default.
- W4214967232 title "Multi-Channel Automatic Speech Recognition Using Deep Complex Unet" @default.
- W4214967232 doi "https://doi.org/10.48550/arxiv.2011.09081" @default.
- W4214967232 hasPublicationYear "2020" @default.
- W4214967232 type Work @default.
- W4214967232 citedByCount "0" @default.
- W4214967232 crossrefType "posted-content" @default.
- W4214967232 hasAuthorship W4214967232A5013004114 @default.
- W4214967232 hasAuthorship W4214967232A5028067869 @default.
- W4214967232 hasAuthorship W4214967232A5029110813 @default.
- W4214967232 hasAuthorship W4214967232A5033976028 @default.
- W4214967232 hasAuthorship W4214967232A5065721838 @default.
- W4214967232 hasAuthorship W4214967232A5077264905 @default.
- W4214967232 hasAuthorship W4214967232A5087994947 @default.
- W4214967232 hasBestOaLocation W42149672321 @default.
- W4214967232 hasConcept C104267543 @default.
- W4214967232 hasConcept C111919701 @default.
- W4214967232 hasConcept C119599485 @default.
- W4214967232 hasConcept C127162648 @default.
- W4214967232 hasConcept C127413603 @default.
- W4214967232 hasConcept C153180895 @default.
- W4214967232 hasConcept C154945302 @default.
- W4214967232 hasConcept C163294075 @default.
- W4214967232 hasConcept C201995342 @default.
- W4214967232 hasConcept C2776182073 @default.
- W4214967232 hasConcept C2778263558 @default.
- W4214967232 hasConcept C2780451532 @default.
- W4214967232 hasConcept C28490314 @default.
- W4214967232 hasConcept C31258907 @default.
- W4214967232 hasConcept C40969351 @default.
- W4214967232 hasConcept C41008148 @default.
- W4214967232 hasConcept C50644808 @default.
- W4214967232 hasConcept C53016008 @default.
- W4214967232 hasConcept C54197355 @default.
- W4214967232 hasConcept C68115822 @default.
- W4214967232 hasConcept C76155785 @default.
- W4214967232 hasConcept C84462506 @default.
- W4214967232 hasConcept C9390403 @default.
- W4214967232 hasConcept C95851461 @default.
- W4214967232 hasConceptScore W4214967232C104267543 @default.
- W4214967232 hasConceptScore W4214967232C111919701 @default.
- W4214967232 hasConceptScore W4214967232C119599485 @default.
- W4214967232 hasConceptScore W4214967232C127162648 @default.
- W4214967232 hasConceptScore W4214967232C127413603 @default.
- W4214967232 hasConceptScore W4214967232C153180895 @default.
- W4214967232 hasConceptScore W4214967232C154945302 @default.
- W4214967232 hasConceptScore W4214967232C163294075 @default.
- W4214967232 hasConceptScore W4214967232C201995342 @default.
- W4214967232 hasConceptScore W4214967232C2776182073 @default.
- W4214967232 hasConceptScore W4214967232C2778263558 @default.
- W4214967232 hasConceptScore W4214967232C2780451532 @default.
- W4214967232 hasConceptScore W4214967232C28490314 @default.
- W4214967232 hasConceptScore W4214967232C31258907 @default.
- W4214967232 hasConceptScore W4214967232C40969351 @default.
- W4214967232 hasConceptScore W4214967232C41008148 @default.
- W4214967232 hasConceptScore W4214967232C50644808 @default.
- W4214967232 hasConceptScore W4214967232C53016008 @default.
- W4214967232 hasConceptScore W4214967232C54197355 @default.
- W4214967232 hasConceptScore W4214967232C68115822 @default.
- W4214967232 hasConceptScore W4214967232C76155785 @default.
- W4214967232 hasConceptScore W4214967232C84462506 @default.
- W4214967232 hasConceptScore W4214967232C9390403 @default.
- W4214967232 hasConceptScore W4214967232C95851461 @default.
- W4214967232 hasLocation W42149672321 @default.
- W4214967232 hasOpenAccess W4214967232 @default.
- W4214967232 hasPrimaryLocation W42149672321 @default.
- W4214967232 hasRelatedWork W1491017269 @default.
- W4214967232 hasRelatedWork W154138180 @default.
- W4214967232 hasRelatedWork W2127095651 @default.
- W4214967232 hasRelatedWork W2130361043 @default.
- W4214967232 hasRelatedWork W2185164075 @default.
- W4214967232 hasRelatedWork W2290075840 @default.
- W4214967232 hasRelatedWork W2337992322 @default.
- W4214967232 hasRelatedWork W2714509385 @default.
- W4214967232 hasRelatedWork W2949767489 @default.
- W4214967232 hasRelatedWork W1954839664 @default.
- W4214967232 isParatext "false" @default.
- W4214967232 isRetracted "false" @default.
- W4214967232 workType "article" @default.