Detailed Information

Cited 0 time in webofscience Cited 0 time in scopus
Metadata Downloads

Deep neural network ensemble for reducing artificial noise in bandwidth extension

Full metadata record
DC Field Value Language
dc.contributor.authorNoh, Kyoungjin-
dc.contributor.authorChang, Joon Hyuk-
dc.date.accessioned2021-07-30T04:54:40Z-
dc.date.available2021-07-30T04:54:40Z-
dc.date.created2021-05-12-
dc.date.issued2020-07-
dc.identifier.issn1051-2004-
dc.identifier.urihttps://scholarworks.bwise.kr/hanyang/handle/2021.sw.hanyang/2036-
dc.description.abstractIn this paper, we propose a deep neural network (DNN) ensemble for reducing artificial noise in speech bandwidth extension (BWE). The proposed DNN ensemble consists of three DNN models; one is a classification model, and the other two are regression models. When estimating sub-band energies of the high-frequency region using sequential DNNs in a frequency domain, the over-estimation of sub-band energies causes annoying artificial noise. To mitigate this artificial noise, we design a DNN classification model that can classify over-estimation frames against normal frames. Then, we separately develop two DNN regression models using half of the entire training set and a limited training set built with overestimation frames and some normal frames to improve the performance at the over-estimation frames. Since the outputs of the classification model are probabilities of either a normal frame or an overestimation frame, respectively, two regression models are adjustably combined by using the probabilistic weights; thus, the final output of the DNN ensemble is the weighted sum of two estimated sub-band energies. As a result, artificial noise is significantly reduced, yielding improved speech quality. The proposed method is objectively and subjectively evaluated by comparing it with conventional approaches.-
dc.language영어-
dc.language.isoen-
dc.publisherACADEMIC PRESS INC ELSEVIER SCIENCE-
dc.titleDeep neural network ensemble for reducing artificial noise in bandwidth extension-
dc.typeArticle-
dc.contributor.affiliatedAuthorChang, Joon Hyuk-
dc.identifier.doi10.1016/j.dsp.2020.102760-
dc.identifier.scopusid2-s2.0-85084373039-
dc.identifier.wosid000536136500008-
dc.identifier.bibliographicCitationDIGITAL SIGNAL PROCESSING, v.102, pp.1 - 6-
dc.relation.isPartOfDIGITAL SIGNAL PROCESSING-
dc.citation.titleDIGITAL SIGNAL PROCESSING-
dc.citation.volume102-
dc.citation.startPage1-
dc.citation.endPage6-
dc.type.rimsART-
dc.type.docTypeArticle-
dc.description.journalClass1-
dc.description.isOpenAccessN-
dc.description.journalRegisteredClassscie-
dc.description.journalRegisteredClassscopus-
dc.relation.journalResearchAreaEngineering-
dc.relation.journalWebOfScienceCategoryEngineering, Electrical & Electronic-
dc.subject.keywordPlusBandwidth-
dc.subject.keywordPlusFrequency domain analysis-
dc.subject.keywordPlusFrequency estimation-
dc.subject.keywordPlusRegression analysis-
dc.subject.keywordPlusDeep neural networks-
dc.subject.keywordPlusArtificial noise-
dc.subject.keywordPlusBandwidth extension-
dc.subject.keywordPlusClassification models-
dc.subject.keywordPlusConventional approach-
dc.subject.keywordPlusFrequency domains-
dc.subject.keywordPlusHigh frequency HF-
dc.subject.keywordPlusNeural network ensembles-
dc.subject.keywordPlusRegression model-
dc.subject.keywordAuthorBandwidth extension-
dc.subject.keywordAuthorDeep neural network-
dc.subject.keywordAuthorEnsemble-
dc.subject.keywordAuthorArtificial noise-
dc.identifier.urlhttps://www.sciencedirect.com/science/article/pii/S1051200420301056?via%3Dihub-
Files in This Item
Go to Link
Appears in
Collections
서울 공과대학 > 서울 융합전자공학부 > 1. Journal Articles

qrcode

Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.

Related Researcher

Researcher Chang, Joon-Hyuk photo

Chang, Joon-Hyuk
COLLEGE OF ENGINEERING (SCHOOL OF ELECTRONIC ENGINEERING)
Read more

Altmetrics

Total Views & Downloads

BROWSE