Phase aware speech enhancement based on depthwise deep convolutional network

Deep neural network-based Speech Enhancement (SE) techniques aim to improve the clarity and quality of speech signals where neural models are trained to recover clean speech from noisy mixtures. Most of these techniques focus on estimating the magnitude of the spectrogram while reusing the phase com...

Full description

Saved in:

Bibliographic Details
Published in:	Signal, image and video processing Vol. 19; no. 9; p. 757
Main Authors:	Iqbal, Yasir, Zhang, Tao, Iqbal, Anjum, Azaz, Ikram, Sadique, Umar, Fahad, Muhammad, Zhao, Xin, Geng, Yanzhang
Format:	Journal Article
Language:	English
Published:	London Springer London 01.09.2025 Springer Nature B.V
Subjects:	Artificial neural networks Computer Imaging Computer Science Datasets Deep learning Distortion Estimation Fourier transforms Image Processing and Computer Vision Intelligibility Machine learning Mapping Multimedia Information Systems Neural networks Noise Original Paper Parallel processing Pattern Recognition and Graphics Signal quality Signal,Image and Speech Processing Software radio Spectrograms Speech Speech processing Vision Speech enhancement Depthwise separable Convolution Signal-to-Distortion ratio (SDR) Complex-Valued spectrograms Depthwise deep convolutional network
ISSN:	1863-1703, 1863-1711
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Abstract	Deep neural network-based Speech Enhancement (SE) techniques aim to improve the clarity and quality of speech signals where neural models are trained to recover clean speech from noisy mixtures. Most of these techniques focus on estimating the magnitude of the spectrogram while reusing the phase component, which limits performance, especially under severe noise distortion. Additionally, high computational demands remain a major challenge for state-of-the-art deep learning models. This paper presents a framework called Depthwise Deep Convolutional Network (DDCN) method for single-channel speech enhancement, leveraging the time-frequency characteristics of speech signals through simultaneous phase and magnitude processing with reduced computations. By incorporating depthwise convolutions in both encoder and decoder architecture the model achieves enhanced computational efficiency. The DDCN SE system adopts parallel processing of the denoised real and imaginary spectrograms which correspond to the magnitude and phase information resulting in superior enhancement performance. The model is trained using a Signal-to-Distortion Ratio (SDR) loss function to minimize distortion and refine phase and magnitude estimation. This ensures that enhanced real and imaginary spectrograms closely match clean speech, improving spectral mapping, and refining the model’s spectral mapping capability. Experimental results demonstrate that the proposed DDCN model, with only 1.28 M parameters, achieves 45.42 FLOPS and RTF of 0.073 outperforms recent deep learning benchmarks, achieving superior speech quality and intelligibility. Specifically, when tested on the WSJ + DNS dataset, the DDCN model outperforms all baselines, achieving a notable SDR improvement of 2.14 dB over the best baseline, CTSNet, across various noise types. Additionally, DDCN surpasses advanced speech enhancement models like ZipEnhancer(S) and MP-SENetUp, achieving WB-PESQ of 3.74 and STOI of 97% on the public VoiceBank + DEMAND dataset.
AbstractList	Deep neural network-based Speech Enhancement (SE) techniques aim to improve the clarity and quality of speech signals where neural models are trained to recover clean speech from noisy mixtures. Most of these techniques focus on estimating the magnitude of the spectrogram while reusing the phase component, which limits performance, especially under severe noise distortion. Additionally, high computational demands remain a major challenge for state-of-the-art deep learning models. This paper presents a framework called Depthwise Deep Convolutional Network (DDCN) method for single-channel speech enhancement, leveraging the time-frequency characteristics of speech signals through simultaneous phase and magnitude processing with reduced computations. By incorporating depthwise convolutions in both encoder and decoder architecture the model achieves enhanced computational efficiency. The DDCN SE system adopts parallel processing of the denoised real and imaginary spectrograms which correspond to the magnitude and phase information resulting in superior enhancement performance. The model is trained using a Signal-to-Distortion Ratio (SDR) loss function to minimize distortion and refine phase and magnitude estimation. This ensures that enhanced real and imaginary spectrograms closely match clean speech, improving spectral mapping, and refining the model’s spectral mapping capability. Experimental results demonstrate that the proposed DDCN model, with only 1.28 M parameters, achieves 45.42 FLOPS and RTF of 0.073 outperforms recent deep learning benchmarks, achieving superior speech quality and intelligibility. Specifically, when tested on the WSJ + DNS dataset, the DDCN model outperforms all baselines, achieving a notable SDR improvement of 2.14 dB over the best baseline, CTSNet, across various noise types. Additionally, DDCN surpasses advanced speech enhancement models like ZipEnhancer(S) and MP-SENetUp, achieving WB-PESQ of 3.74 and STOI of 97% on the public VoiceBank + DEMAND dataset. Deep neural network-based Speech Enhancement (SE) techniques aim to improve the clarity and quality of speech signals where neural models are trained to recover clean speech from noisy mixtures. Most of these techniques focus on estimating the magnitude of the spectrogram while reusing the phase component, which limits performance, especially under severe noise distortion. Additionally, high computational demands remain a major challenge for state-of-the-art deep learning models. This paper presents a framework called Depthwise Deep Convolutional Network (DDCN) method for single-channel speech enhancement, leveraging the time-frequency characteristics of speech signals through simultaneous phase and magnitude processing with reduced computations. By incorporating depthwise convolutions in both encoder and decoder architecture the model achieves enhanced computational efficiency. The DDCN SE system adopts parallel processing of the denoised real and imaginary spectrograms which correspond to the magnitude and phase information resulting in superior enhancement performance. The model is trained using a Signal-to-Distortion Ratio (SDR) loss function to minimize distortion and refine phase and magnitude estimation. This ensures that enhanced real and imaginary spectrograms closely match clean speech, improving spectral mapping, and refining the model’s spectral mapping capability. Experimental results demonstrate that the proposed DDCN model, with only 1.28 M parameters, achieves 45.42 FLOPS and RTF of 0.073 outperforms recent deep learning benchmarks, achieving superior speech quality and intelligibility. Specifically, when tested on the WSJ + DNS dataset, the DDCN model outperforms all baselines, achieving a notable SDR improvement of 2.14 dB over the best baseline, CTSNet, across various noise types. Additionally, DDCN surpasses advanced speech enhancement models like ZipEnhancer(S) and MP-SENetUp, achieving WB-PESQ of 3.74 and STOI of 97% on the public VoiceBank + DEMAND dataset.
ArticleNumber	757
Author	Fahad, Muhammad Azaz, Ikram Iqbal, Anjum Zhao, Xin Iqbal, Yasir Geng, Yanzhang Zhang, Tao Sadique, Umar
Author_xml	– sequence: 1 givenname: Yasir surname: Iqbal fullname: Iqbal, Yasir organization: School of Electrical and Information Engineering, Tianjin University – sequence: 2 givenname: Tao surname: Zhang fullname: Zhang, Tao organization: School of Electrical and Information Engineering, Tianjin University – sequence: 3 givenname: Anjum surname: Iqbal fullname: Iqbal, Anjum organization: School of Software Technology, Dalian University of Technology – sequence: 4 givenname: Ikram surname: Azaz fullname: Azaz, Ikram organization: School of Electrical and Information Engineering, Tianjin University – sequence: 5 givenname: Umar surname: Sadique fullname: Sadique, Umar organization: School of Electrical and Information Engineering, Tianjin University – sequence: 6 givenname: Muhammad surname: Fahad fullname: Fahad, Muhammad organization: School of Electrical and Information Engineering, Tianjin University – sequence: 7 givenname: Xin surname: Zhao fullname: Zhao, Xin organization: School of Electrical and Information Engineering, Tianjin University – sequence: 8 givenname: Yanzhang surname: Geng fullname: Geng, Yanzhang email: gregory@tju.edu.cn organization: School of Electrical and Information Engineering, Tianjin University
BookMark	eNp9kE1LAzEQhoMoWGv_gKeA5-gk2d1sj1L8goIe9Byy2Ynb2iZrsrX4742u6M25ZCDP-zI8J-TQB4-EnHG44ADqMnGuKmAgSgaFlMD4AZnwupKMK84Pf3eQx2SW0hrySKHqqp6Q5WNnElKzNxFp6hFtR9F3xlvcoh9ok39bGjxtsR-6_SqzLWJPbfDvYbMbVsGbDfU47EN8PSVHzmwSzn7eKXm-uX5a3LHlw-394mrJrFBiYChRNLIBZ6q6laWqXYUCeDGfGyhQWiuMqOZNzcE1TevQAGa0sI6XrpQtyCk5H3v7GN52mAa9DruYD0lailLlqgJUpsRI2RhSiuh0H1dbEz80B_0lTo_idBanv8VpnkNyDKUM-xeMf9X_pD4BhX1y8Q
Cites_doi	10.1007/s11277-021-08313-6 10.1142/s0219467825500019 10.1007/s10115-022-01818-x 10.1109/ICASSP49660.2025.10888703 10.21437/Interspeech.2021-2207 10.1109/ICASSP40776.2020.9053188 10.1109/TASSP.1978.1163086 10.1109/AISP57993.2023.10134933 10.1109/ICASSP49357.2023.10096208 10.32604/iasc.2023.028090 10.1609/aaai.v34i05.6489 10.1121/1.4799597 10.1109/ICASSP39728.2021.9414177 10.1109/ICASSP43922.2022.9747120 10.1177/23312165231209913 10.1016/j.apacoust.2023.109385 10.1007/s11760-024-03500-x 10.1109/ICASSP39728.2021.9414878 10.1109/TASSP.1985.1164550 10.3233/IDT-230211 10.21437/Interspeech.2020-2537 10.3115/1075527.1075614 10.1016/j.specom.2010.12.003 10.1109/TASLP.2019.2955276 10.1016/j.dsp.2024.104408 10.24963/ijcai.2022/582 10.1109/ICASSP39728.2021.9414062 10.1109/ICSDA.2013.6709856 10.1007/s00034-023-02455-7 10.1109/ICPR56361.2022.9956638 10.21437/Interspeech.2020-2143 10.1121/1.5055562 10.1109/ICASSP.2019.8682834 10.1016/j.neunet.2025.107562 10.1109/ICASSP49660.2025.10890034 10.21437/Interspeech.2021-1609 10.1109/TASSP.1979.1163209 10.1016/0167-6393(93)90095-3 10.1142/S0219467825500676 10.1109/TASLP.2021.3079813 10.1109/LSP.2025.3558690 10.1016/j.apacoust.2023.109592 10.3991/ijoe.v19i04.37577 10.1016/j.specom.2023.103008 10.21437/SSW.2016-24
ContentType	Journal Article
Copyright	The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.
Copyright_xml	– notice: The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025 Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. – notice: The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.
DBID	AAYXX CITATION JQ2
DOI	10.1007/s11760-025-04330-1
DatabaseName	CrossRef ProQuest Computer Science Collection
DatabaseTitle	CrossRef ProQuest Computer Science Collection
DatabaseTitleList	ProQuest Computer Science Collection
DeliveryMethod	fulltext_linktorsrc
Discipline	Engineering Computer Science
EISSN	1863-1711
ExternalDocumentID	10_1007_s11760_025_04330_1
GrantInformation_xml	– fundername: National Natural Science Foundation of China grantid: 62271344 funderid: https://doi.org/10.13039/501100001809
GroupedDBID	.VR 06D 0R~ 123 1N0 203 29~ 2J2 2JN 2JY 2KG 2KM 2LR 2~H 30V 4.4 406 408 409 40D 40E 5VS 67Z 6NX 875 8TC 95- 95. 95~ AAAVM AABHQ AACDK AAHNG AAIAL AAJBT AAJKR AANZL AAPKM AARTL AASML AATNV AATVU AAUYE AAWCG AAYIU AAYQN AAYTO AAYZH ABAKF ABBRH ABBXA ABDBE ABDZT ABECU ABFSG ABFTV ABHQN ABJNI ABJOX ABKCH ABMNI ABMQK ABNWP ABQBU ABSXP ABTEG ABTHY ABTKH ABTMW ABWNU ABXPI ACAOD ACDTI ACGFS ACHSB ACHXU ACKNC ACMDZ ACMLO ACOKC ACOMO ACPIV ACSNA ACSTC ACZOJ ADHHG ADHIR ADKFA ADKNI ADKPE ADRFC ADTPH ADURQ ADYFF ADZKW AEFQL AEGAL AEGNC AEJHL AEJRE AEMSY AENEX AEOHA AEPYU AESKC AETLH AEVLU AEXYK AEZWR AFBBN AFDZB AFHIU AFLOW AFOHR AFQWF AFWTZ AFZKB AGAYW AGDGC AGMZJ AGQEE AGQMX AGRTI AGWIL AGWZB AGYKE AHAVH AHBYD AHPBZ AHWEU AHYZX AIGIU AIIXL AILAN AITGF AIXLP AJRNO AJZVZ ALMA_UNASSIGNED_HOLDINGS ALWAN AMKLP AMXSW AMYLF AMYQR AOCGG ARMRJ ATHPR AXYYD AYFIA AYJHY B-. BA0 BGNMA BSONS CS3 CSCUP DDRTE DNIVK DPUIP EBLON EBS EIOEI ESBYG FERAY FFXSO FIGPU FNLPD FRRFC FWDCC GGCAI GGRSB GJIRD GNWQR GQ7 GQ8 GXS HF~ HG5 HG6 HLICF HMJXF HQYDN HRMNR HZ~ IJ- IKXTQ IWAJR IXC IXD IXE IZIGR IZQ I~X I~Z J-C J0Z JBSCW JCJTX JZLTJ KDC KOV LLZTM M4Y MA- NPVJJ NQJWS NU0 O93 O9J OAM P9O PF0 PT4 QOS R89 R9I ROL RPX RSV S16 S1Z S27 S3B SAP SDH SEG SHX SISQX SJYHP SNE SNPRN SNX SOHCF SOJ SPISZ SRMVM SSLCW STPWE SZN T13 TSG TSK TSV TUC U2A UG4 UOJIU UTJUX UZXMN VC2 VFIZW W48 YLTOR Z45 ZMTXR ~A9 -Y2 2VQ AARHV AAYXX ABRTQ ABULA ACBXY AEBTG AFFHD AFGCZ AFKRA AGJBK AHSBF AIAKS AJBLW ARAPS BDATZ BENPR BGLVJ CAG CCPQU CITATION COF EJD FINBP FSGXE H13 HCIFZ K7- O9- PHGZM PHGZT PQGLB JQ2
ID	FETCH-LOGICAL-c272t-e3e2b3b0fa68d3578f6e201499a04e3cc2a269b810fbbdfea0efa64cf15f53d03
IEDL.DBID	RSV
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001512019700003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	1863-1703
IngestDate	Sat Oct 11 06:55:55 EDT 2025 Sat Nov 29 07:47:04 EST 2025 Fri Jul 04 01:22:12 EDT 2025
IsPeerReviewed	true
IsScholarly	true
Issue	9
Keywords	Speech enhancement Depthwise separable Convolution Signal-to-Distortion ratio (SDR) Complex-Valued spectrograms Depthwise deep convolutional network
Language	English
LinkModel	DirectLink
MergedId	FETCHMERGED-LOGICAL-c272t-e3e2b3b0fa68d3578f6e201499a04e3cc2a269b810fbbdfea0efa64cf15f53d03
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 14
PQID	3257014407
PQPubID	2044169
ParticipantIDs	proquest_journals_3257014407 crossref_primary_10_1007_s11760_025_04330_1 springer_journals_10_1007_s11760_025_04330_1
PublicationCentury	2000
PublicationDate	2025-09-01
PublicationDateYYYYMMDD	2025-09-01
PublicationDate_xml	– month: 09 year: 2025 text: 2025-09-01 day: 01
PublicationDecade	2020
PublicationPlace	London
PublicationPlace_xml	– name: London – name: Heidelberg
PublicationTitle	Signal, image and video processing
PublicationTitleAbbrev	SIViP
PublicationYear	2025
Publisher	Springer London Springer Nature B.V
Publisher_xml	– name: Springer London – name: Springer Nature B.V
References	J Lim (4330_CR4) 1978; 26 C Jannu (4330_CR27) 2023; 42 4330_CR41 4330_CR1 4330_CR40 S Sivapatham (4330_CR14) 2023; 212 C Zheng (4330_CR43) 2023; 27 4330_CR45 4330_CR44 4330_CR42 4330_CR49 4330_CR47 4330_CR46 C Jannu (4330_CR7) 2025; 25 N Saleem (4330_CR12) 2024; 147 4330_CR30 Y Zhao (4330_CR48) 2018; 144 4330_CR34 4330_CR33 4330_CR32 4330_CR31 4330_CR38 4330_CR37 4330_CR36 4330_CR35 C Jannu (4330_CR28) 2023; 45 FE Wahab (4330_CR25) 2024; 156 A Karthik (4330_CR13) 2021; 119 A Li (4330_CR24) 2021; 29 4330_CR23 4330_CR22 4330_CR21 4330_CR20 4330_CR26 V Parisae (4330_CR29) 2024; 46 H Guo (4330_CR39) 2023; 209 RR Rai (4330_CR9) 2024; 18 4330_CR50 4330_CR5 4330_CR11 4330_CR10 4330_CR3 4330_CR8 4330_CR16 4330_CR15 4330_CR6 4330_CR19 4330_CR18 K Tan (4330_CR2) 2019; 28 4330_CR17
References_xml	– volume: 119 start-page: 1959 issue: 3 year: 2021 ident: 4330_CR13 publication-title: Wireless Pers. Commun. doi: 10.1007/s11277-021-08313-6 – volume: 25 start-page: 2550001 issue: 01 year: 2025 ident: 4330_CR7 publication-title: Int. J. Image Graphics doi: 10.1142/s0219467825500019 – ident: 4330_CR31 doi: 10.1007/s10115-022-01818-x – ident: 4330_CR49 doi: 10.1109/ICASSP49660.2025.10888703 – ident: 4330_CR1 doi: 10.21437/Interspeech.2021-2207 – ident: 4330_CR6 doi: 10.1109/ICASSP40776.2020.9053188 – volume: 26 start-page: 197 issue: 3 year: 1978 ident: 4330_CR4 publication-title: IEEE Trans. Acoust. Speech Signal Process. doi: 10.1109/TASSP.1978.1163086 – ident: 4330_CR26 doi: 10.1109/AISP57993.2023.10134933 – ident: 4330_CR20 doi: 10.1109/ICASSP49357.2023.10096208 – ident: 4330_CR21 – ident: 4330_CR11 doi: 10.32604/iasc.2023.028090 – ident: 4330_CR19 doi: 10.1609/aaai.v34i05.6489 – ident: 4330_CR35 doi: 10.1121/1.4799597 – ident: 4330_CR44 doi: 10.1109/ICASSP39728.2021.9414177 – ident: 4330_CR40 doi: 10.1109/ICASSP43922.2022.9747120 – volume: 27 start-page: 233121652312099 year: 2023 ident: 4330_CR43 publication-title: Trends Hear. doi: 10.1177/23312165231209913 – volume: 209 start-page: 109385 year: 2023 ident: 4330_CR39 publication-title: Appl. Acoust. doi: 10.1016/j.apacoust.2023.109385 – ident: 4330_CR18 doi: 10.1007/s11760-024-03500-x – ident: 4330_CR47 doi: 10.1109/ICASSP39728.2021.9414878 – ident: 4330_CR22 – ident: 4330_CR5 doi: 10.1109/TASSP.1985.1164550 – volume: 18 start-page: 123 year: 2024 ident: 4330_CR9 publication-title: Intell. Decis. Technol. doi: 10.3233/IDT-230211 – ident: 4330_CR17 doi: 10.21437/Interspeech.2020-2537 – ident: 4330_CR36 doi: 10.3115/1075527.1075614 – ident: 4330_CR16 doi: 10.1016/j.specom.2010.12.003 – volume: 28 start-page: 380 year: 2019 ident: 4330_CR2 publication-title: IEEE/ACM Trans. Audio Speech Lang. Process. doi: 10.1109/TASLP.2019.2955276 – volume: 147 start-page: 104408 year: 2024 ident: 4330_CR12 publication-title: Digit. Signal Proc. doi: 10.1016/j.dsp.2024.104408 – ident: 4330_CR46 doi: 10.24963/ijcai.2022/582 – ident: 4330_CR45 doi: 10.1109/ICASSP39728.2021.9414062 – ident: 4330_CR34 doi: 10.1109/ICSDA.2013.6709856 – volume: 42 start-page: 7467 issue: 12 year: 2023 ident: 4330_CR27 publication-title: Circuits Syst. Signal. Process. doi: 10.1007/s00034-023-02455-7 – volume: 46 start-page: 10907 issue: 4 year: 2024 ident: 4330_CR29 publication-title: J. Intell. Fuzzy Syst. – ident: 4330_CR32 doi: 10.1109/ICPR56361.2022.9956638 – ident: 4330_CR8 doi: 10.21437/Interspeech.2020-2143 – volume: 144 start-page: 1627 issue: 3 year: 2018 ident: 4330_CR48 publication-title: J. Acoust. Soc. Am. doi: 10.1121/1.5055562 – ident: 4330_CR23 doi: 10.1109/ICASSP.2019.8682834 – ident: 4330_CR50 doi: 10.1016/j.neunet.2025.107562 – ident: 4330_CR41 doi: 10.1109/ICASSP49660.2025.10890034 – ident: 4330_CR37 doi: 10.21437/Interspeech.2021-1609 – ident: 4330_CR3 doi: 10.1109/TASSP.1979.1163209 – ident: 4330_CR38 doi: 10.1016/0167-6393(93)90095-3 – ident: 4330_CR10 – ident: 4330_CR30 doi: 10.1142/S0219467825500676 – volume: 29 start-page: 1829 year: 2021 ident: 4330_CR24 publication-title: IEEE/ACM Trans. Audio Speech Lang. Process. doi: 10.1109/TASLP.2021.3079813 – ident: 4330_CR42 doi: 10.1109/LSP.2025.3558690 – volume: 212 start-page: 109592 year: 2023 ident: 4330_CR14 publication-title: Appl. Acoust. doi: 10.1016/j.apacoust.2023.109592 – ident: 4330_CR15 doi: 10.3991/ijoe.v19i04.37577 – volume: 156 start-page: 103008 year: 2024 ident: 4330_CR25 publication-title: Speech Commun. doi: 10.1016/j.specom.2023.103008 – volume: 45 start-page: 1195 issue: 1 year: 2023 ident: 4330_CR28 publication-title: J. Intell. Fuzzy Syst. – ident: 4330_CR33 doi: 10.21437/SSW.2016-24
SSID	ssj0000327868
Score	2.343979
Snippet	Deep neural network-based Speech Enhancement (SE) techniques aim to improve the clarity and quality of speech signals where neural models are trained to...
SourceID	proquest crossref springer
SourceType	Aggregation Database Index Database Publisher
StartPage	757
SubjectTerms	Artificial neural networks Computer Imaging Computer Science Datasets Deep learning Distortion Estimation Fourier transforms Image Processing and Computer Vision Intelligibility Machine learning Mapping Multimedia Information Systems Neural networks Noise Original Paper Parallel processing Pattern Recognition and Graphics Signal quality Signal,Image and Speech Processing Software radio Spectrograms Speech Speech processing Vision
Title	Phase aware speech enhancement based on depthwise deep convolutional network
URI	https://link.springer.com/article/10.1007/s11760-025-04330-1 https://www.proquest.com/docview/3257014407
Volume	19
WOSCitedRecordID	wos001512019700003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText	1
inHoldings	1
isFullTextHit
isPrint
journalDatabaseRights	– providerCode: PRVAVX databaseName: SpringerLINK Contemporary 1997-Present customDbUrl: eissn: 1863-1711 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000327868 issn: 1863-1703 databaseCode: RSV dateStart: 20070401 isFulltext: true titleUrlDefault: https://link.springer.com/search?facet-content-type=%22Journal%22 providerName: Springer Nature
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV1LT8MwDLbQ4AAHBgPEYKAcuEGkNOma9ogQEwc0Tby0W5Wkjsalm9YBf5-kDwoIDnCu5VaOXTuJ_X0AZ9YVqUpyTYcJMhrywFKPOkYlMwID6yJK6JJsQo7H8XSaTOqhsKLpdm-uJMs_dTvsFsiIUU-_6kG3GHV7nnWX7mJP2HB3__RxssIEl3E1AxdHHn-TiXpa5mc1XzNSW2Z-uxktE86o-79P3YHtusAkl5VH7MIa5j3oNuQNpI7lHmx9QiLcg9vJzKUzot7UEkmxQDQzgvnMu4Q_PiQ-2WVknpMMF55-3clmiAvim9Zr53Uvzaue8n14HF0_XN3QmmiBGi75iqJAroVmVkVx5uFvbITc750SxUIUxnDFo0THAbNaZxYVQycaGhsM7VBkTBxAJ5_neAjEWJRoeZhZEYeJcAqslRg5c6ggQm36cN4YO11UeBppi5zszZY6s6Wl2dKgD4NmPdI6topUeOI9fyct-3DR2L99_Lu2o7-JH8MmL5fQN5QNoLNavuAJbJjX1XOxPC197h0HP9Ew
linkProvider	Springer Nature
linkToHtml	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwnV3BTsMwDLXQQAIODAaIwYAcuEGkNOma9ogQ0xBjmmCg3ao2dTQuXbUO-H2SrqWA4ADnWm7l2LWT2O8BnGlTpEaSx7QbIKMudzS1qGNUMiXQ0SaiRFyQTcjh0J9MglE5FJZX3e7VlWTxp66H3RzpMWrpVy3oFqNmz7NqlHctYv79w9PHyQoTXPrLGTjfs_ibTJTTMj-r-ZqR6jLz281okXB6zf996jZslQUmuVx6xA6sYNqCZkXeQMpYbsHmJyTCXRiMpiadkegtmiPJM0Q1JZhOrUvY40Nik11CZilJMLP060Y2QcyIbVovnde8NF32lO_BY-96fNWnJdECVVzyBUWBPBYx05HnJxb-RnvI7d4piJiLQikecS-IfYfpOE40RgyNqKu009VdkTCxD410luIBEKVRouZuooXvBsIo0FqiZ8wROR7Gqg3nlbHDbImnEdbIydZsoTFbWJgtdNrQqdYjLGMrD4Ul3rN30rINF5X968e_azv8m_gprPfHd4NwcDO8PYINXiynbS7rQGMxf8FjWFOvi-d8flL43zve39QU
linkToPdf	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwpV1LT8MwDLYQIAQH3ojxzIEbRKRJ17RHBEwgpmkSD3Gr2tTRuHTVNuDvE_fBBoID4lzLqRxbtmP7M8CJdUFqomXK2xEK7kvPckId41oYhZ51FqXSctmE7vXC5-eoPzPFX3a7NyXJaqaBUJryyXmR2fPp4JunA8FpFSsBcAnu8p8FnxrpKV-_f_p8ZRFK6rCahwsDwuIUqp6c-ZnNV-80DTm_VUlL59NZ-_9vr8NqHXiyi0pTNmAO801Ya5Y6sNrGN2FlBqFwC7r9gXNzLHlPRsjGBaIZMMwHpCr0rMjICWZsmLMMC1rL7mgzxIJRM3ut1O7QvOo134bHzvXD5Q2vFzBwI7WccFQoU5UKmwRhRrA4NkBJOVWUCB-VMTKRQZSGnrBpmllMBDpS31ivbdsqE2oH5vNhjrvAjEWNVvqZVaEfKcfAWo2BE0fiBZiaFpw2go-LCmcjniIqk9hiJ7a4FFvsteCguZu4trlxrGghH9WqdQvOmruYfv6d297fyI9hqX_Vibu3vbt9WJblbVLP2QHMT0aveAiL5m3yMh4dlar4AeV93Pg
openUrl	ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Phase+aware+speech+enhancement+based+on+depthwise+deep+convolutional+network&rft.jtitle=Signal%2C+image+and+video+processing&rft.au=Iqbal%2C+Yasir&rft.au=Zhang%2C+Tao&rft.au=Iqbal%2C+Anjum&rft.au=Azaz%2C+Ikram&rft.date=2025-09-01&rft.pub=Springer+London&rft.issn=1863-1703&rft.eissn=1863-1711&rft.volume=19&rft.issue=9&rft_id=info:doi/10.1007%2Fs11760-025-04330-1&rft.externalDocID=10_1007_s11760_025_04330_1
thumbnail_l	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1863-1703&client=summon
thumbnail_m	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1863-1703&client=summon
thumbnail_s	http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1863-1703&client=summon