A cross-linguistic depression detection method based on speech data

Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting thei...

Celý popis

Uložené v:

Podrobná bibliografia
Vydané v:	Journal of affective disorders Ročník 390; s. 119739
Hlavní autori:	Qin, Shengjie, Zhang, Yuezhou, Ma, Yuliang, Li, Hui, Li, Xingxing, Lian, Bin, Cai, Weiming, Cui, Jialin, Zhao, Xianghong
Médium:	Journal Article
Jazyk:	English
Vydavateľské údaje:	Netherlands Elsevier B.V 01.12.2025
Predmet:	Adult Artificial Intelligence Cross-linguistic Deep Learning Depression Depression - diagnosis Female Humans Language Linguistics Male Psychiatric/Mental Health Speech Transfer learning Depression Speech Cross-linguistic Transfer learning
ISSN:	0165-0327, 1573-2517, 1573-2517
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!

Abstract	Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization. We propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models. Our model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression. Future research should incorporate a wider range of languages. These findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions. [Display omitted] •This study demonstrates the feasibility of cross-linguistic depression detection using speech data.•Significant speech feature differences exist within Chinese/English groups (healthy vs depressed) and cross-linguistically.•The proposed method aligns cross-lingual speech features, streamlining multilingual dataset construction.•The proposed method achieves accuracies of 88.7% (English, source domain) and 81.1% (Chinese, target domain).
AbstractList	AbstractBackgroundDepression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization. MethodsWe propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models. ResultsOur model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression. LimitationsFuture research should incorporate a wider range of languages. ConclusionsThese findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions. Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization.BACKGROUNDDepression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization.We propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models.METHODSWe propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models.Our model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression.RESULTSOur model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression.Future research should incorporate a wider range of languages.LIMITATIONSFuture research should incorporate a wider range of languages.These findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions.CONCLUSIONSThese findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions. Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization. We propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models. Our model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression. Future research should incorporate a wider range of languages. These findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions. [Display omitted] •This study demonstrates the feasibility of cross-linguistic depression detection using speech data.•Significant speech feature differences exist within Chinese/English groups (healthy vs depressed) and cross-linguistically.•The proposed method aligns cross-lingual speech features, streamlining multilingual dataset construction.•The proposed method achieves accuracies of 88.7% (English, source domain) and 81.1% (Chinese, target domain). Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence, researchers have focused on automated depression detection using speech data. However, most current AI methods rely on monolingual data, limiting their cross-linguistic generalization. We propose a transfer learning method called Deep Covariance Alignment Network (DCAN) that transfers models trained on English speech data (source domain) to Chinese speech data (target domain). Experiments were conducted on the DAIC-WOZ and MODMA datasets. We used down-sampled speech data (1 kHz), features extracted by a Convolutional AutoEncoder, and manually selected features to explore commonalities across languages and compare our method with other models. Our model achieved an accuracy of 88.7 % on the English dataset and 81.1 % on the Chinese dataset, outperforming models trained solely on English data by an average of 21.9 %. Compared to other mainstream transfer learning methods, our approach showed a 4 % improvement. This reveals that even across different linguistic and cultural backgrounds, there are potential commonalities between speech signals and depression. Future research should incorporate a wider range of languages. These findings highlight that our research enhances the generalization capability of depression detection models based on speech data across different linguistic domains, thereby reducing the effort required for constructing diverse language-specific speech datasets. The method we proposed are expected to be useful in supporting the diagnosis of depressions.
ArticleNumber	119739
Author	Lian, Bin Cai, Weiming Qin, Shengjie Ma, Yuliang Li, Xingxing Li, Hui Zhang, Yuezhou Zhao, Xianghong Cui, Jialin
Author_xml	– sequence: 1 givenname: Shengjie surname: Qin fullname: Qin, Shengjie organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China – sequence: 2 givenname: Yuezhou surname: Zhang fullname: Zhang, Yuezhou organization: Department of Biostatistics & Health Informatics, Institute of Psychiatry, Psychology and Neuroscience, King's College London, London, United Kingdom – sequence: 3 givenname: Yuliang surname: Ma fullname: Ma, Yuliang organization: Intelligent Control & Robotics Institute, College of Automation, Hangzhou Dianzi University, Hangzhou 310000, China – sequence: 4 givenname: Hui surname: Li fullname: Li, Hui organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China – sequence: 5 givenname: Xingxing surname: Li fullname: Li, Xingxing organization: The Affiliated Kangning Hospital of Ningbo University, NingBo 315100, China – sequence: 6 givenname: Bin surname: Lian fullname: Lian, Bin organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China – sequence: 7 givenname: Weiming surname: Cai fullname: Cai, Weiming organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China – sequence: 8 givenname: Jialin surname: Cui fullname: Cui, Jialin organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China – sequence: 9 givenname: Xianghong surname: Zhao fullname: Zhao, Xianghong email: zxh@nit.net.cn organization: School of Information Science and Engineering, NingboTech University, Ningbo 315100, China
BackLink	https://www.ncbi.nlm.nih.gov/pubmed/40570935$$D View this record in MEDLINE/PubMed
BookMark	eNqFkUtLxDAUhYOM6MzoD3AjXbrpmGc7RRCGwRcILtR1SJMbTe20Y9IK8-9NHXUhqKvcwDknOd-doFHTNoDQEcEzgkl2Ws0qZWYUUzEjpMhZsYPGROQspYLkIzSOGpFiRvN9NAmhwhhnRY730D7HIscFE2O0XCTatyGktWueehc6pxMDaw8huLaJYwe6G6YVdM-tSUoVwCTxHtYA-jkxqlMHaNeqOsDh5zlFj5cXD8vr9Pbu6ma5uE01J7RLVVFolumSZ5SpaLZzbYDTkltrNJ_r0iquhSixLTNjFYsfhJJZYSCzuaE5m6KTbe7at689hE6uXNBQ16qBtg-SUcozNhdUROnxp7QvV2Dk2ruV8hv5VTwKyFbw0d6D_ZYQLAe4spIRrhzgyi3c6DnbeiCWfHPgZdAOGg3G-UhJmtb96T7_4daRudOqfoENhKrtfRPpSSIDlVjeD8sbdkcFJmROeAwofg_45_F3VUGpqA
Cites_doi	10.1016/j.specom.2024.103106 10.1001/archpsyc.1981.01780370046005 10.1016/j.bspc.2023.105898 10.1016/j.bspc.2023.105675 10.1016/j.bspc.2021.103107 10.1109/ACCESS.2019.2936124 10.1002/mpr.256 10.1016/S0165-0327(99)00020-8 10.3390/app8081372 10.1109/10.846676 10.1136/bmj.297.6653.897 10.1038/mp.2010.13 10.1037/0022-006X.64.6.1417 10.1007/s12046-011-0050-4 10.1046/j.1525-1497.2001.016009606.x 10.3390/cancers13040738 10.1016/j.jad.2016.10.019 10.1016/j.specom.2015.03.004 10.1186/s40537-016-0043-6 10.1016/j.bbe.2021.11.004 10.1016/j.jad.2023.08.097 10.1016/j.jad.2022.11.060 10.1016/j.jad.2024.03.106 10.1016/j.jad.2024.08.002 10.1016/j.jad.2021.08.090 10.1016/j.eswa.2021.116076
ContentType	Journal Article
Copyright	2025 Elsevier B.V. Elsevier B.V. Copyright © 2025 Elsevier B.V. All rights reserved.
Copyright_xml	– notice: 2025 Elsevier B.V. – notice: Elsevier B.V. – notice: Copyright © 2025 Elsevier B.V. All rights reserved.
DBID	AAYXX CITATION CGR CUY CVF ECM EIF NPM 7X8
DOI	10.1016/j.jad.2025.119739
DatabaseName	CrossRef Medline MEDLINE MEDLINE (Ovid) MEDLINE MEDLINE PubMed MEDLINE - Academic
DatabaseTitle	CrossRef MEDLINE Medline Complete MEDLINE with Full Text PubMed MEDLINE (Ovid) MEDLINE - Academic
DatabaseTitleList	MEDLINE - Academic MEDLINE
Database_xml	– sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database
DeliveryMethod	fulltext_linktorsrc
EISSN	1573-2517
EndPage	119739
ExternalDocumentID	40570935 10_1016_j_jad_2025_119739 S0165032725011814 1_s2_0_S0165032725011814
Genre	Journal Article
GroupedDBID	--- --K --M .1- .FO .~1 0R~ 1B1 1P~ 1RT 1~. 1~5 4.4 457 4G. 5GY 5VS 7-5 71M 8P~ 9JM AABNK AAEDW AAIKJ AAKOC AALRI AAOAW AAQFI AATTM AAWTL AAXKI AAXUO ABBQC ABFNM ABIVO ABJNI ABLJU ABMAC ABMZM ACDAQ ACGFS ACHQT ACIEU ACIUM ACLOT ACRLP ACVFH ADBBV ADCNI ADEZE AEBSH AEIPS AEKER AENEX AEUPX AEVXI AFJKZ AFPUW AFRHN AFTJW AFXIZ AGUBO AGYEJ AHHHB AIEXJ AIGII AIIUN AIKHN AITUG AJRQY AJUYK AKBMS AKRWK AKYEP ALMA_UNASSIGNED_HOLDINGS AMRAJ ANKPU ANZVX APXCP AXJTR BKOJK BLXMC BNPGV CS3 DU5 EBS EFJIC EFKBS EFLBG EO8 EO9 EP2 EP3 F5P FDB FIRID FNPLU FYGXN G-Q GBLVA HMQ HMW IHE J1W KOM M29 M2V M39 M3V M41 MO0 N9A O-L O9- OAUVE OH0 OU- OZT P-8 P-9 P2P PC. Q38 ROL RPZ SAE SCC SDF SDG SDP SEL SES SEW SPCBC SSH SSZ T5K UV1 Z5R ~G- ~HD 29J 53G AAEDT AAGKA AAQXK AAYWO ABWVN ABXDB ACRPL ADMUD ADNMO ADVLN AGHFR AGQPQ ASPBG AVWKF AZFZN EJD FEDTE FGOYB G-2 HEG HMK HMO HVGLF HZ~ R2- SNS SPS WUQ ZGI 9DU AAYXX CITATION AGCQF CGR CUY CVF ECM EIF NPM 7X8
ID	FETCH-LOGICAL-c412t-a99c36cb4623aeecf8cde42b4ffdc48cbfa4c55b0fb6dfa3093eb3f5de6f7d273
ISICitedReferencesCount	0
ISICitedReferencesURI	http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001527393400003&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN	0165-0327 1573-2517
IngestDate	Sat Nov 01 15:02:44 EDT 2025 Sat Sep 06 11:16:20 EDT 2025 Thu Nov 27 00:34:51 EST 2025 Wed Dec 10 14:22:58 EST 2025 Sun Oct 19 01:21:16 EDT 2025 Wed Dec 10 14:10:28 EST 2025
IsPeerReviewed	true
IsScholarly	true
Keywords	Depression Speech Cross-linguistic Transfer learning
Language	English
License	Copyright © 2025 Elsevier B.V. All rights reserved.
LinkModel	OpenURL
MergedId	FETCHMERGED-LOGICAL-c412t-a99c36cb4623aeecf8cde42b4ffdc48cbfa4c55b0fb6dfa3093eb3f5de6f7d273
Notes	ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23
PMID	40570935
PQID	3224638525
PQPubID	23479
PageCount	1
ParticipantIDs	proquest_miscellaneous_3224638525 pubmed_primary_40570935 crossref_primary_10_1016_j_jad_2025_119739 elsevier_sciencedirect_doi_10_1016_j_jad_2025_119739 elsevier_clinicalkeyesjournals_1_s2_0_S0165032725011814 elsevier_clinicalkey_doi_10_1016_j_jad_2025_119739
PublicationCentury	2000
PublicationDate	2025-12-01
PublicationDateYYYYMMDD	2025-12-01
PublicationDate_xml	– month: 12 year: 2025 text: 2025-12-01 day: 01
PublicationDecade	2020
PublicationPlace	Netherlands
PublicationPlace_xml	– name: Netherlands
PublicationTitle	Journal of affective disorders
PublicationTitleAlternate	J Affect Disord
PublicationYear	2025
Publisher	Elsevier B.V
Publisher_xml	– name: Elsevier B.V
References	Du, Liu, Wang (bb0050) 2023; 323 Rejaibi, Komaty, Meriaudeau (bb0120) 2022; 71 Ayana, Dese, Choe (bb0010) 2021; 13 Penninx, Beekman, Smit (bb0110) 2008; 17 Wang, Ravi, Flint (bb0165) 2024; 163 Zhou, He, Xu (bb0190) 2018; 8 Cai, Gao, Sun (bb0025) 2020 Kroenke, Spitzer, Williams (bb0095) 2001; 16 France, Shiavi, Silverman (bb5000) 2000; 47 Cui, Wang, Zhuo (bb0030) 2020 Taşcı (bb0150) 2024; 364 Ross, Rush (bb0130) 1981; 38 Fried (bb0055) 2017; 208 Goldberg, Bridges, Duncan-Jones (bb0065) 1988; 297 Rosenquist, Fowler, Christakis (bb0125) 2011; 16 Bourlard, Dines, Magimai-Doss (bb0020) 2011; 36 Klein, Schatzberg, McCullough (bb0085) 1999; 55 Gratch, Artstein, Lucas (bb5005) 2014; 14 Kora, Ooi, Faust (bb0090) 2022; 42 Van der Maaten, Hinton (bb0160) 2008; 9 Zhang, Folarin, Dineley (bb0185) 2024; 355 Yu, Wang, Chen (bb0180) 2019 Khalil, Jones, Babar (bb0080) 2019; 7 Houlsby, Giurgiu, Jastrzebski (bb0075) 2019 Das, Naskar (bb0045) 2024; 90 Sun, Jiang, Gao (bb0145) 2024; 88 Tian, Zhu, Jing (bb0155) 2023 Weiss, Khoshgoftaar, Wang (bb0170) 2016; 3 Cummins, Dineley, Conde (bb0040) 2023; 341 Ma, Yang, Chen (bb0100) 2016 Cummins, Scherer, Krajewski (bb0035) 2015; 71 Senoussaoui, Sarria-Paja, Santos (bb0140) 2014 He, Zhang, Ren (bb0070) 2016 Mitra, Shriberg, McLaren (bb0105) 2014 Ye, Yu, Wang (bb0175) 2021; 295 Beck, Steer, Brown (bb0015) 1996 Ganin, Ustinova, Ajakan (bb0060) 2016; 17 Addis, Jacobson (bb0005) 1996; 64 Sardari, Nakisa, Rastgoo (bb0135) 2022; 189 Polzehl, Möller, Metze (bb0115) 2010 Fried (10.1016/j.jad.2025.119739_bb0055) 2017; 208 Zhang (10.1016/j.jad.2025.119739_bb0185) 2024; 355 Das (10.1016/j.jad.2025.119739_bb0045) 2024; 90 Mitra (10.1016/j.jad.2025.119739_bb0105) 2014 Sun (10.1016/j.jad.2025.119739_bb0145) 2024; 88 He (10.1016/j.jad.2025.119739_bb0070) 2016 Kora (10.1016/j.jad.2025.119739_bb0090) 2022; 42 Rosenquist (10.1016/j.jad.2025.119739_bb0125) 2011; 16 Weiss (10.1016/j.jad.2025.119739_bb0170) 2016; 3 Ayana (10.1016/j.jad.2025.119739_bb0010) 2021; 13 Gratch (10.1016/j.jad.2025.119739_bb5005) 2014; 14 Beck (10.1016/j.jad.2025.119739_bb0015) 1996 France (10.1016/j.jad.2025.119739_bb5000) 2000; 47 Rejaibi (10.1016/j.jad.2025.119739_bb0120) 2022; 71 Zhou (10.1016/j.jad.2025.119739_bb0190) 2018; 8 Houlsby (10.1016/j.jad.2025.119739_bb0075) 2019 Cummins (10.1016/j.jad.2025.119739_bb0040) 2023; 341 Khalil (10.1016/j.jad.2025.119739_bb0080) 2019; 7 Klein (10.1016/j.jad.2025.119739_bb0085) 1999; 55 Ganin (10.1016/j.jad.2025.119739_bb0060) 2016; 17 Ma (10.1016/j.jad.2025.119739_bb0100) 2016 Senoussaoui (10.1016/j.jad.2025.119739_bb0140) 2014 Sardari (10.1016/j.jad.2025.119739_bb0135) 2022; 189 Yu (10.1016/j.jad.2025.119739_bb0180) 2019 Cui (10.1016/j.jad.2025.119739_bb0030) 2020 Ross (10.1016/j.jad.2025.119739_bb0130) 1981; 38 Tian (10.1016/j.jad.2025.119739_bb0155) 2023 Polzehl (10.1016/j.jad.2025.119739_bb0115) 2010 Cai (10.1016/j.jad.2025.119739_bb0025) 2020 Cummins (10.1016/j.jad.2025.119739_bb0035) 2015; 71 Goldberg (10.1016/j.jad.2025.119739_bb0065) 1988; 297 Taşcı (10.1016/j.jad.2025.119739_bb0150) 2024; 364 Addis (10.1016/j.jad.2025.119739_bb0005) 1996; 64 Ye (10.1016/j.jad.2025.119739_bb0175) 2021; 295 Penninx (10.1016/j.jad.2025.119739_bb0110) 2008; 17 Du (10.1016/j.jad.2025.119739_bb0050) 2023; 323 Bourlard (10.1016/j.jad.2025.119739_bb0020) 2011; 36 Wang (10.1016/j.jad.2025.119739_bb0165) 2024; 163 Kroenke (10.1016/j.jad.2025.119739_bb0095) 2001; 16 Van der Maaten (10.1016/j.jad.2025.119739_bb0160) 2008; 9
References_xml	– volume: 189 year: 2022 ident: bb0135 article-title: Audio based depression detection using convolutional Autoencoder publication-title: Expert Syst. Appl. – volume: 341 start-page: 128 year: 2023 end-page: 136 ident: bb0040 article-title: Multilingual markers of depression in remotely collected speech samples: a preliminary analysis publication-title: J. Affect. Disord. – volume: 13 start-page: 738 year: 2021 ident: bb0010 article-title: Transfer learning in breast cancer diagnoses via ultrasound imaging publication-title: Cancers – volume: 47 start-page: 829 year: 2000 end-page: 837 ident: bb5000 article-title: Acoustical properties of speech as indicators of depression and suicidal risk publication-title: IEEE Trans. Biomed. Eng. – volume: 71 year: 2022 ident: bb0120 article-title: MFCC-based recurrent neural network for automatic clinical depression recognition and assessment from speech publication-title: Biomed. Signal Process. Control. – volume: 355 start-page: 40 year: 2024 end-page: 49 ident: bb0185 article-title: Identifying depression-related topics in smartphone-collected free-response speech recordings using an automatic speech recognition system and a deep learning topic model publication-title: J. Affect. Disord. – volume: 71 start-page: 10 year: 2015 end-page: 49 ident: bb0035 article-title: A review of depression and suicide risk assessment using speech analysis publication-title: Speech Comm. – volume: 38 start-page: 1344 year: 1981 end-page: 1354 ident: bb0130 article-title: Diagnosis and neuroanatomical correlates of depression in brain-damaged patients: implications for a neurology of depression publication-title: Arch. Gen. Psychiatry – volume: 8 start-page: 1372 year: 2018 ident: bb0190 article-title: Feature selection and transfer learning for Alzheimers disease clinical diagnosis publication-title: Appl. Sci. – volume: 7 start-page: 117327 year: 2019 end-page: 117345 ident: bb0080 article-title: Speech emotion recognition using deep learning techniques: a review publication-title: IEEE access. – volume: 16 start-page: 273 year: 2011 end-page: 281 ident: bb0125 article-title: Social network determinants of depression publication-title: Mol. Psychiatry – start-page: 2790 year: 2019 end-page: 2799 ident: bb0075 article-title: Parameter-efficient transfer learning for NLP publication-title: International Conference on Machine Learning. PMLR – start-page: 57 year: 2014 end-page: 63 ident: bb0140 article-title: Model fusion for multimodal depression classification and level detection publication-title: Proceedings of the 4th International Workshop on Audio/Visual Emotion Challenge – volume: 295 start-page: 904 year: 2021 end-page: 913 ident: bb0175 article-title: Multi-modal depression detection based on emotional audio and evaluation text publication-title: J. Affect. Disord. – volume: 36 start-page: 885 year: 2011 end-page: 915 ident: bb0020 article-title: Current trends in multilingual speech processing publication-title: Sadhana – volume: 3 start-page: 1 year: 2016 end-page: 40 ident: bb0170 article-title: A survey of transfer learning publication-title: J. Big Data – volume: 208 start-page: 191 year: 2017 end-page: 197 ident: bb0055 article-title: The 52 symptoms of major depression: lack of content overlap among seven common depression scales publication-title: J. Affect. Disord. – start-page: 1 year: 2023 end-page: 16 ident: bb0155 article-title: Deep learning for depression recognition from speech publication-title: Mobile Networks and Applications. (online) – volume: 323 start-page: 299 year: 2023 end-page: 308 ident: bb0050 article-title: Depression recognition using a proposed speech chain model fusing speech production and perception features publication-title: J. Affect. Disord. – volume: 90 year: 2024 ident: bb0045 article-title: A deep learning model for depression detection based on MFCC and CNN generated spectrogram features publication-title: Biomed. Signal Process. Control – start-page: 35 year: 2016 end-page: 42 ident: bb0100 article-title: Depaudionet: an efficient deep model for audio based depression classification publication-title: Proceedings of the 6th International Workshop on Audio/Visual Emotion Challenge – start-page: 93 year: 2014 end-page: 101 ident: bb0105 article-title: The SRI AVEC-2014 evaluation system publication-title: Proceedings of the 4th International Workshop on Audio/Visual Emotion Challenge – volume: 88 year: 2024 ident: bb0145 article-title: A novel study for depression detecting using audio signals based on graph neural network publication-title: Biomed. Signal Process. Control. – start-page: 134 year: 2010 end-page: 140 ident: bb0115 article-title: Automatically assessing personality from speech publication-title: 2010 IEEE Fourth International Conference on Semantic Computing. IEEE – volume: 9 year: 2008 ident: bb0160 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. – volume: 163 year: 2024 ident: bb0165 article-title: Speechformer-ctc: sequential modeling of depression detection with speech temporal classification publication-title: Speech Commun. – volume: 14 start-page: 3123 year: 2014 end-page: 3128 ident: bb5005 article-title: The distress analysis interview corpus of human and computer interviews[C]//LREC – volume: 16 start-page: 606 year: 2001 end-page: 613 ident: bb0095 article-title: The PHQ-9: validity of a brief depression severity measure publication-title: J. Gen. Intern. Med. – volume: 55 start-page: 149 year: 1999 end-page: 157 ident: bb0085 article-title: Age of onset in chronic major depression: relation to demographic and clinical variables, family history, and treatment response publication-title: J. Affect. Disord. – year: 2020 ident: bb0025 article-title: Modma Dataset: A Multi-modal Open Dataset for Mental-disorder Analysis. arXiv preprint arXiv.2002.09283 – start-page: 3941 year: 2020 end-page: 3950 ident: bb0030 article-title: Towards discriminability and diversity: batch nuclear-norm maximization under label insufficient situations publication-title: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition – volume: 17 start-page: 121 year: 2008 end-page: 140 ident: bb0110 article-title: The Netherlands study of depression and anxiety (NESDA): rationale, objectives and methods publication-title: Int. J. Methods Psychiatr. Res. – start-page: 778 year: 2019 end-page: 786 ident: bb0180 article-title: Transfer learning with dynamic adversarial adaptation network publication-title: 2019 IEEE International Conference on Data Mining. IEEE – volume: 64 start-page: 1417 year: 1996 ident: bb0005 article-title: Reasons for depression and the process and outcome of cognitive–behavioral psychotherapies publication-title: J. Consult. Clin. Psychol. – year: 1996 ident: bb0015 article-title: Beck Depression Inventory. – start-page: 770 year: 2016 end-page: 778 ident: bb0070 article-title: Deep residual learning for image recognition publication-title: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition – volume: 297 start-page: 897 year: 1988 end-page: 899 ident: bb0065 article-title: Detecting anxiety and depression in general medical settings publication-title: Br. Med. J. – volume: 364 start-page: 9 year: 2024 end-page: 19 ident: bb0150 article-title: Multilevel hybrid handcrafted feature extraction based depression recognition method using speech publication-title: J. Affect. Disord. – volume: 42 start-page: 79 year: 2022 end-page: 107 ident: bb0090 article-title: Transfer learning techniques for medical image analysis: a review publication-title: Biocybernetics Biomed. Eng. – volume: 17 start-page: 1 year: 2016 end-page: 35 ident: bb0060 article-title: Domain-adversarial training of neural networks publication-title: J. Mach. Learn. Res. – volume: 163 year: 2024 ident: 10.1016/j.jad.2025.119739_bb0165 article-title: Speechformer-ctc: sequential modeling of depression detection with speech temporal classification publication-title: Speech Commun. doi: 10.1016/j.specom.2024.103106 – volume: 17 start-page: 1 year: 2016 ident: 10.1016/j.jad.2025.119739_bb0060 article-title: Domain-adversarial training of neural networks publication-title: J. Mach. Learn. Res. – volume: 38 start-page: 1344 year: 1981 ident: 10.1016/j.jad.2025.119739_bb0130 article-title: Diagnosis and neuroanatomical correlates of depression in brain-damaged patients: implications for a neurology of depression publication-title: Arch. Gen. Psychiatry doi: 10.1001/archpsyc.1981.01780370046005 – volume: 90 year: 2024 ident: 10.1016/j.jad.2025.119739_bb0045 article-title: A deep learning model for depression detection based on MFCC and CNN generated spectrogram features publication-title: Biomed. Signal Process. Control doi: 10.1016/j.bspc.2023.105898 – year: 2020 ident: 10.1016/j.jad.2025.119739_bb0025 – volume: 14 start-page: 3123 year: 2014 ident: 10.1016/j.jad.2025.119739_bb5005 article-title: The distress analysis interview corpus of human and computer interviews[C]//LREC – volume: 88 year: 2024 ident: 10.1016/j.jad.2025.119739_bb0145 article-title: A novel study for depression detecting using audio signals based on graph neural network publication-title: Biomed. Signal Process. Control. doi: 10.1016/j.bspc.2023.105675 – volume: 71 year: 2022 ident: 10.1016/j.jad.2025.119739_bb0120 article-title: MFCC-based recurrent neural network for automatic clinical depression recognition and assessment from speech publication-title: Biomed. Signal Process. Control. doi: 10.1016/j.bspc.2021.103107 – volume: 9 issue: 11 year: 2008 ident: 10.1016/j.jad.2025.119739_bb0160 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. – volume: 7 start-page: 117327 year: 2019 ident: 10.1016/j.jad.2025.119739_bb0080 article-title: Speech emotion recognition using deep learning techniques: a review publication-title: IEEE access. doi: 10.1109/ACCESS.2019.2936124 – start-page: 93 year: 2014 ident: 10.1016/j.jad.2025.119739_bb0105 article-title: The SRI AVEC-2014 evaluation system – volume: 17 start-page: 121 year: 2008 ident: 10.1016/j.jad.2025.119739_bb0110 article-title: The Netherlands study of depression and anxiety (NESDA): rationale, objectives and methods publication-title: Int. J. Methods Psychiatr. Res. doi: 10.1002/mpr.256 – volume: 55 start-page: 149 year: 1999 ident: 10.1016/j.jad.2025.119739_bb0085 article-title: Age of onset in chronic major depression: relation to demographic and clinical variables, family history, and treatment response publication-title: J. Affect. Disord. doi: 10.1016/S0165-0327(99)00020-8 – year: 1996 ident: 10.1016/j.jad.2025.119739_bb0015 – start-page: 778 year: 2019 ident: 10.1016/j.jad.2025.119739_bb0180 article-title: Transfer learning with dynamic adversarial adaptation network – volume: 8 start-page: 1372 year: 2018 ident: 10.1016/j.jad.2025.119739_bb0190 article-title: Feature selection and transfer learning for Alzheimers disease clinical diagnosis publication-title: Appl. Sci. doi: 10.3390/app8081372 – volume: 47 start-page: 829 year: 2000 ident: 10.1016/j.jad.2025.119739_bb5000 article-title: Acoustical properties of speech as indicators of depression and suicidal risk publication-title: IEEE Trans. Biomed. Eng. doi: 10.1109/10.846676 – volume: 297 start-page: 897 year: 1988 ident: 10.1016/j.jad.2025.119739_bb0065 article-title: Detecting anxiety and depression in general medical settings publication-title: Br. Med. J. doi: 10.1136/bmj.297.6653.897 – volume: 16 start-page: 273 year: 2011 ident: 10.1016/j.jad.2025.119739_bb0125 article-title: Social network determinants of depression publication-title: Mol. Psychiatry doi: 10.1038/mp.2010.13 – volume: 64 start-page: 1417 year: 1996 ident: 10.1016/j.jad.2025.119739_bb0005 article-title: Reasons for depression and the process and outcome of cognitive–behavioral psychotherapies publication-title: J. Consult. Clin. Psychol. doi: 10.1037/0022-006X.64.6.1417 – volume: 36 start-page: 885 year: 2011 ident: 10.1016/j.jad.2025.119739_bb0020 article-title: Current trends in multilingual speech processing publication-title: Sadhana doi: 10.1007/s12046-011-0050-4 – start-page: 57 year: 2014 ident: 10.1016/j.jad.2025.119739_bb0140 article-title: Model fusion for multimodal depression classification and level detection – start-page: 1 year: 2023 ident: 10.1016/j.jad.2025.119739_bb0155 article-title: Deep learning for depression recognition from speech – volume: 16 start-page: 606 year: 2001 ident: 10.1016/j.jad.2025.119739_bb0095 article-title: The PHQ-9: validity of a brief depression severity measure publication-title: J. Gen. Intern. Med. doi: 10.1046/j.1525-1497.2001.016009606.x – volume: 13 start-page: 738 year: 2021 ident: 10.1016/j.jad.2025.119739_bb0010 article-title: Transfer learning in breast cancer diagnoses via ultrasound imaging publication-title: Cancers doi: 10.3390/cancers13040738 – volume: 208 start-page: 191 year: 2017 ident: 10.1016/j.jad.2025.119739_bb0055 article-title: The 52 symptoms of major depression: lack of content overlap among seven common depression scales publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2016.10.019 – volume: 71 start-page: 10 year: 2015 ident: 10.1016/j.jad.2025.119739_bb0035 article-title: A review of depression and suicide risk assessment using speech analysis publication-title: Speech Comm. doi: 10.1016/j.specom.2015.03.004 – volume: 3 start-page: 1 year: 2016 ident: 10.1016/j.jad.2025.119739_bb0170 article-title: A survey of transfer learning publication-title: J. Big Data doi: 10.1186/s40537-016-0043-6 – start-page: 770 year: 2016 ident: 10.1016/j.jad.2025.119739_bb0070 article-title: Deep residual learning for image recognition – volume: 42 start-page: 79 year: 2022 ident: 10.1016/j.jad.2025.119739_bb0090 article-title: Transfer learning techniques for medical image analysis: a review publication-title: Biocybernetics Biomed. Eng. doi: 10.1016/j.bbe.2021.11.004 – start-page: 2790 year: 2019 ident: 10.1016/j.jad.2025.119739_bb0075 article-title: Parameter-efficient transfer learning for NLP – volume: 341 start-page: 128 year: 2023 ident: 10.1016/j.jad.2025.119739_bb0040 article-title: Multilingual markers of depression in remotely collected speech samples: a preliminary analysis publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2023.08.097 – start-page: 35 year: 2016 ident: 10.1016/j.jad.2025.119739_bb0100 article-title: Depaudionet: an efficient deep model for audio based depression classification – start-page: 3941 year: 2020 ident: 10.1016/j.jad.2025.119739_bb0030 article-title: Towards discriminability and diversity: batch nuclear-norm maximization under label insufficient situations – volume: 323 start-page: 299 year: 2023 ident: 10.1016/j.jad.2025.119739_bb0050 article-title: Depression recognition using a proposed speech chain model fusing speech production and perception features publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2022.11.060 – start-page: 134 year: 2010 ident: 10.1016/j.jad.2025.119739_bb0115 article-title: Automatically assessing personality from speech – volume: 355 start-page: 40 year: 2024 ident: 10.1016/j.jad.2025.119739_bb0185 article-title: Identifying depression-related topics in smartphone-collected free-response speech recordings using an automatic speech recognition system and a deep learning topic model publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2024.03.106 – volume: 364 start-page: 9 year: 2024 ident: 10.1016/j.jad.2025.119739_bb0150 article-title: Multilevel hybrid handcrafted feature extraction based depression recognition method using speech publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2024.08.002 – volume: 295 start-page: 904 year: 2021 ident: 10.1016/j.jad.2025.119739_bb0175 article-title: Multi-modal depression detection based on emotional audio and evaluation text publication-title: J. Affect. Disord. doi: 10.1016/j.jad.2021.08.090 – volume: 189 year: 2022 ident: 10.1016/j.jad.2025.119739_bb0135 article-title: Audio based depression detection using convolutional Autoencoder publication-title: Expert Syst. Appl. doi: 10.1016/j.eswa.2021.116076
SSID	ssj0006970
Score	2.484237
Snippet	Depression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial intelligence,... AbstractBackgroundDepression is a common and disabling psychological disorder that affects patients and their social circles. With advances in artificial...
SourceID	proquest pubmed crossref elsevier
SourceType	Aggregation Database Index Database Publisher
StartPage	119739
SubjectTerms	Adult Artificial Intelligence Cross-linguistic Deep Learning Depression Depression - diagnosis Female Humans Language Linguistics Male Psychiatric/Mental Health Speech Transfer learning
Title	A cross-linguistic depression detection method based on speech data
URI	https://www.clinicalkey.com/#!/content/1-s2.0-S0165032725011814 https://www.clinicalkey.es/playcontent/1-s2.0-S0165032725011814 https://dx.doi.org/10.1016/j.jad.2025.119739 https://www.ncbi.nlm.nih.gov/pubmed/40570935 https://www.proquest.com/docview/3224638525
Volume	390
WOSCitedRecordID	wos001527393400003&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: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1573-2517 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0006970 issn: 0165-0327 databaseCode: AIEXJ dateStart: 20150101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier
link	http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV1Li9swEBbZbA-9lJa-0sfiQk8NCo4sWdIxhJS-WFp2C-lJ2LK0SShOWMfL0mN_eTWWH-nupi_oxSSDlMiaT5rRaB4IvQyFFZmkFhOZckytNDhJOMNxqMeRZXFGuK2KTfDjYzGfy4-93vcmFubiK89zcXkpN_-V1Y7mmA2hs3_B7vZHHcF9dkx3T8d29_wjxk-GleTDEGdeVmmYO3dX8HzdGl8d3NeOHoIYy-DKoNgYoxfDOljtJo01qXw_wNUoq3N2tgr5J5-K4GRh8rPVsgVLa47-Uppvi3XZ2b89EWwsZw3xgy-gXS53LRGE7Xh11MbJmOEw8rH-ze4a-Wqg9f4Il5Y-eVEtbTvCtb3cmxVWo1UCGV0JG-123s2bPZu-H-OCjEJ8AkOAETjFDiJq6c-N_alHFUSF6lrTA3RIOJOijw4nb2fzd608j2VVcrB9u-ZuvPISvDK4fdrNvtNLpcWc3kV3amYGEw-be6hn8vtoOgmuQiboIBO0kAk8ZIIKMoH77iETAGQeoM-vZ6fTN7guroE1HZMtTqTUUaxT6vTfxDW2QmeGkpRam2kqdGoTqhlLQ5vGmU3gwtykbvVmJrbcreDoIern69w8RgETMrU0hVRxgpoUKihwoblOxpQbK-wAvWomRW18DhXVOBeulJtBBTOo_AwOEGmmTTXBwU6cKYeKX3XiN3UyRb0uC7WP6wNE2561zul1yd_94YuGo8rtx3DJluRmXRYqggyNkWCEDdAjz-r2peFwBI4HT_55uE_R7W7lPUP97XlpnqNb-mK7LM6P0AGfi6Mavz8A7jC3TQ
linkProvider	Elsevier
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=A+cross-linguistic+depression+detection+method+based+on+speech+data&rft.jtitle=Journal+of+affective+disorders&rft.au=Qin%2C+Shengjie&rft.au=Zhang%2C+Yuezhou&rft.au=Ma%2C+Yuliang&rft.au=Li%2C+Hui&rft.date=2025-12-01&rft.issn=0165-0327&rft.volume=390&rft.spage=119739&rft.epage=119739&rft_id=info:doi/10.1016%2Fj.jad.2025.119739&rft.externalDBID=ECK1-s2.0-S0165032725011814&rft.externalDocID=1_s2_0_S0165032725011814
thumbnail_m	http://cvtisr.summon.serialssolutions.com/2.0.0/image/custom?url=https%3A%2F%2Fcdn.clinicalkey.com%2Fck-thumbnails%2F01650327%2FS0165032725X00185%2Fcov150h.gif