{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,4]],"date-time":"2026-09-04T20:16:50Z","timestamp":1788553010765,"version":"build-2803163510"},"reference-count":72,"publisher":"Wiley","issue":"2","license":[{"start":{"date-parts":[[2022,9,18]],"date-time":"2022-09-18T00:00:00Z","timestamp":1663459200000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/onlinelibrary.wiley.com\/termsAndConditions#vor"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["51875221(J.G.)"],"award-info":[{"award-number":["51875221(J.G.)"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Intl J Lang &amp;amp; Comm Disor"],"published-print":{"date-parts":[[2023,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:sec><jats:title>Background<\/jats:title><jats:p>Auditory\u2013perceptual assessment of voice is a subjective procedure. Artificial intelligence with deep learning (DL) may improve the consistency and accessibility of this task. It is unclear how a DL model performs on different acoustic features.<\/jats:p><\/jats:sec><jats:sec><jats:title>Aims<\/jats:title><jats:p>To develop a generalizable DL framework for identifying dysphonia using a multidimensional acoustic feature.<\/jats:p><\/jats:sec><jats:sec><jats:title>Methods &amp; Procedures<\/jats:title><jats:p>Recordings of sustained phonations of \/a\/ and \/i\/ were retrospectively collected from a clinical database. Subjects contained 238 dysphonic and 223 vocally healthy speakers of Chinese Mandarin. All audio clips were split into multiple 1.5\u2010s segments and normalized to the same loudness level. Mel frequency cepstral coefficients and mel\u2010spectrogram were extracted from these standardized segments. Each set of features was used in a convolutional neural network (CNN) to perform a binary classification task. The best feature was obtained through a five\u2010fold cross\u2010validation on a random selection of 80% data. The resultant DL framework was tested on the remaining 20% data and a public German voice database. The performance of the DL framework was compared with those of two baseline machine\u2010learning models.<\/jats:p><\/jats:sec><jats:sec><jats:title>Outcomes &amp; Results<\/jats:title><jats:p>The mel\u2010spectrogram yielded the best model performance, with a mean area under the receiver operating characteristic curve of 0.972 and an accuracy of 92% in classifying audio segments. The resultant DL framework significantly outperformed both baseline models in detecting dysphonic subjects on both test sets. The best outcomes were achieved when classifications were made based on all segments of both vowels, with 95% accuracy, 92% recall, 98% precision and 98% specificity on the Chinese test set, and 92%, 95%, 90% and 89%, respectively, on the German set.<\/jats:p><\/jats:sec><jats:sec><jats:title>Conclusions &amp; Implications<\/jats:title><jats:p>This study demonstrates the feasibility of DL for automatic detection of dysphonia. The mel\u2010spectrogram is a preferred acoustic feature for the task. This framework may be used for vocal health screening and facilitate automatic perceptual evaluation of voice in the era of big data.<\/jats:p><\/jats:sec><jats:sec><jats:title>WHAT THIS PAPER ADDS<\/jats:title><jats:sec><jats:title>What is already known on this subject<\/jats:title><jats:p><jats:list list-type=\"bullet\">\n<jats:list-item><jats:p>Auditory\u2013perceptual assessment is the current gold standard in clinical evaluation of voice quality, but its value may be limited by the rater's reliability and accessibility. DL is a new method of artificial intelligence that can overcome these disadvantages and promote automatic voice assessment. This study explored the feasibility of a DL approach for automatic detection of dysphonia, along with a quantitative comparison of two common sets of acoustic features.<\/jats:p><\/jats:list-item><\/jats:list><\/jats:p><\/jats:sec><jats:sec><jats:title>What this study adds to existing knowledge<\/jats:title><jats:p><jats:list list-type=\"bullet\">\n<jats:list-item><jats:p>A CNN model is excellent at decoding multidimensional acoustic features, outperforming the baseline parameter\u2010based models in identifying dysphonic voices. The first 13 mel\u2010frequency cepstral coefficients (MFCCs) are sufficient for this task. The mel\u2010spectrogram results in greater performance, indicating the acoustic features are presented in a more favourable way than the MFCCs to the CNN model.<\/jats:p><\/jats:list-item>\n<\/jats:list><\/jats:p><\/jats:sec><jats:sec><jats:title>What are the potential or actual clinical implications of this work?<\/jats:title><jats:p><jats:list list-type=\"bullet\">\n<jats:list-item><jats:p>DL is a feasible method for the detection of dysphonia. The current DL framework may be used for remote vocal health screening or documenting voice recovery after treatment. In future, DL models may potentially be used to perform auditory\u2013perceptual tasks in an automatic, efficient, reliable and low\u2010cost manner.<\/jats:p><\/jats:list-item>\n<\/jats:list><\/jats:p><\/jats:sec><\/jats:sec>","DOI":"10.1111\/1460-6984.12783","type":"journal-article","created":{"date-parts":[[2022,9,19]],"date-time":"2022-09-19T06:04:04Z","timestamp":1663567444000},"page":"279-294","update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":30,"title":["Deep learning in automatic detection of dysphonia: Comparing acoustic features and developing a generalizable framework"],"prefix":"10.1111","volume":"58","author":[{"given":"Zhen","family":"Chen","sequence":"first","affiliation":[{"name":"Department of Rehabilitation Sciences East China Normal University  Shanghai China"},{"name":"Department of Otolaryngology\u2014Head &amp; Neck Surgery Eye, Ear, Nose and Throat Hospital, Fudan University  Shanghai China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Peixi","family":"Zhu","sequence":"additional","affiliation":[{"name":"Hilderbrand Department of Petroleum and Geosystems Engineering University of Texas at Austin  Austin TX USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wei","family":"Qiu","sequence":"additional","affiliation":[{"name":"Hangzhou Chenqing Heye Technology Co., Ltd  Hangzhou Zhejiang China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiajie","family":"Guo","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Digital Manufacturing Equipment and Technology School of Mechanical Science and Engineering Huazhong University of Science and Technology  Wuhan Hubei China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8465-130X","authenticated-orcid":false,"given":"Yike","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Otolaryngology\u2014Head and Neck Surgery Vanderbilt University Medical Center  Nashville TN USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"311","published-online":{"date-parts":[[2022,9,18]]},"reference":[{"key":"e_1_2_10_2_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2016.09.009"},{"key":"e_1_2_10_3_1","doi-asserted-by":"publisher","DOI":"10.1007\/S10916\u2010015\u20100392\u20102\/FIGURES\/6"},{"key":"e_1_2_10_4_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.JVOICE.2015.08.010"},{"key":"e_1_2_10_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2017.2680467"},{"key":"e_1_2_10_6_1","doi-asserted-by":"publisher","DOI":"10.5772\/intechopen.80419"},{"key":"e_1_2_10_7_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.JVOICE.2016.01.014"},{"key":"e_1_2_10_8_1","unstructured":"American Speech\u2013Language\u2013Hearing Association. (2021)Voice disorders.https:\/\/www.asha.org\/practice\u2010portal\/clinical\u2010topics\/voice\u2010disorders\/"},{"key":"e_1_2_10_9_1","doi-asserted-by":"crossref","unstructured":"An K.H. Kim M. Teplansky K. Green J.R. Campbell T.F. Yunusova Y. et\u00a0al. (2018)Automatic early detection of amyotrophic lateral sclerosis from intelligible speech using convolutional neural networks.Proceedings of the Annual Conference of the International Speech Communication Association INTERSPEECH 2018\u2010September 1913\u20131917.https:\/\/doi.org\/10.21437\/Interspeech.2018\u20102496","DOI":"10.21437\/Interspeech.2018-2496"},{"key":"e_1_2_10_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/TBME.2010.2089052"},{"key":"e_1_2_10_11_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.anl.2014.11.001"},{"key":"e_1_2_10_12_1","doi-asserted-by":"publisher","DOI":"10.1001\/jamaoto.2018.0980"},{"issue":"9","key":"e_1_2_10_13_1","first-page":"341","article-title":"Praat, a system for doing phonetics by computer","volume":"5","author":"Boersma P","year":"2002","journal-title":"Glot International"},{"key":"e_1_2_10_14_1","doi-asserted-by":"publisher","DOI":"10.1002\/LARY.29595"},{"key":"e_1_2_10_15_1","doi-asserted-by":"publisher","DOI":"10.1177\/000348940611500209"},{"key":"e_1_2_10_16_1","doi-asserted-by":"publisher","DOI":"10.1002\/lary.23189"},{"key":"e_1_2_10_17_1","doi-asserted-by":"publisher","DOI":"10.1002\/lary.24280"},{"key":"e_1_2_10_18_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.amjmed.2014.10.040"},{"key":"e_1_2_10_19_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591\u2010018\u20100177\u20105"},{"key":"e_1_2_10_20_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2013.10.022"},{"key":"e_1_2_10_21_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.ymeth.2018.07.007"},{"key":"e_1_2_10_22_1","doi-asserted-by":"publisher","DOI":"10.1109\/TASSP.1980.1163420"},{"key":"e_1_2_10_23_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591\u2010018\u20100107\u20106"},{"key":"e_1_2_10_24_1","doi-asserted-by":"publisher","DOI":"10.2307\/2531595"},{"key":"e_1_2_10_25_1","doi-asserted-by":"publisher","DOI":"10.1155\/2015\/956249"},{"key":"e_1_2_10_26_1","doi-asserted-by":"publisher","DOI":"10.1038\/s41591\u2010018\u20100316\u2010z"},{"key":"e_1_2_10_27_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2018.02.003"},{"key":"e_1_2_10_28_1","doi-asserted-by":"publisher","DOI":"10.1044\/2016_JSLHR\u2010S\u201015\u20100307"},{"key":"e_1_2_10_29_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0144610"},{"key":"e_1_2_10_30_1","doi-asserted-by":"crossref","unstructured":"Hammami I. Salhi L.&Labidi S.(2016)Pathological voices detection using Support Vector Machine. 2nd International Conference on Advanced Technologies for Signal and Image Processing ATSIP 2016 662\u2013666.https:\/\/doi.org\/10.1109\/ATSIP.2016.7523162","DOI":"10.1109\/ATSIP.2016.7523162"},{"key":"e_1_2_10_31_1","doi-asserted-by":"crossref","unstructured":"Han W. Chan C.F. Choy C.S.&Pun K.P.(2006)An efficient MFCC extraction method in speech recognition.Proceedings\u2014IEEE International Symposium on Circuits and Systems 145\u2013148.https:\/\/doi.org\/10.1109\/iscas.2006.1692543","DOI":"10.1109\/ISCAS.2006.1692543"},{"key":"e_1_2_10_32_1","unstructured":"Han Y.&Lee K.(2016)Acoustic scene classification using convolutional neural network and multiple\u2010width frequency\u2010delta data augmentation.http:\/\/arxiv.org\/abs\/1607.02383"},{"key":"e_1_2_10_33_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2018.07.014"},{"key":"e_1_2_10_34_1","volume-title":"Clinical examination of voice","author":"Hirano M.","year":"1981"},{"key":"e_1_2_10_35_1","doi-asserted-by":"publisher","DOI":"10.1002\/lary.26947"},{"key":"e_1_2_10_36_1","first-page":"1","volume-title":"Applied computing in medicine and health","author":"Iram S.","year":"2015"},{"key":"e_1_2_10_37_1","doi-asserted-by":"crossref","unstructured":"Juvela L. Bollepalli B. Wang X. Kameoka H. Airaksinen M. Yamagishi J. et\u00a0al. (2018)Speech Waveform Synthesis from MFCC Sequences with Generative Adversarial Networks.ICASSP IEEE International Conference on Acoustics Speech and Signal Processing\u2014Proceedings 2018 April 5679\u20135683.https:\/\/doi.org\/10.1109\/ICASSP.2018.8461852","DOI":"10.1109\/ICASSP.2018.8461852"},{"key":"e_1_2_10_38_1","doi-asserted-by":"publisher","DOI":"10.1186\/s12916\u2010019\u20101426\u20102"},{"key":"e_1_2_10_39_1","unstructured":"Kingma D. P.&Ba J.(2015)Adam: a method for stochastic optimization.International Conference on Learning Representations.http:\/\/arxiv.org\/abs\/1412.6980"},{"key":"e_1_2_10_40_1","doi-asserted-by":"crossref","unstructured":"Kinnunen T.&Alku P.(2009)On separating glottal source and vocal tract information in telephony speaker verification.ICASSP IEEE International Conference on Acoustics Speech and Signal Processing\u2014Proceedings 4545\u20134548.https:\/\/doi.org\/10.1109\/ICASSP.2009.4960641","DOI":"10.1109\/ICASSP.2009.4960641"},{"key":"e_1_2_10_41_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0095001"},{"key":"e_1_2_10_42_1","doi-asserted-by":"publisher","DOI":"10.1044\/jshr.3601.21"},{"key":"e_1_2_10_43_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2004.01.007"},{"key":"e_1_2_10_44_1","doi-asserted-by":"publisher","DOI":"10.1038\/nature14539"},{"key":"e_1_2_10_45_1","doi-asserted-by":"publisher","DOI":"10.1155\/2013\/758731"},{"key":"e_1_2_10_46_1","doi-asserted-by":"publisher","DOI":"10.3233\/JAD\u2010210714"},{"key":"e_1_2_10_47_1","unstructured":"librosa: Audio and Music Signal Analysis in Python (2015)PROC. OF THE 14th PYTHON IN SCIENCE CONF.http:\/\/conference.scipy.org\/proceedings\/scipy2015\/pdfs\/brian_mcfee.pdf"},{"key":"e_1_2_10_48_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2015.09.017"},{"key":"e_1_2_10_49_1","doi-asserted-by":"publisher","DOI":"10.1097\/MOO.0b013e3282fe96ce"},{"key":"e_1_2_10_50_1","doi-asserted-by":"publisher","DOI":"10.1155\/2017\/8783751"},{"key":"e_1_2_10_51_1","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.2017.1600425CM"},{"key":"e_1_2_10_52_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.JVOICE.2012.03.005"},{"key":"e_1_2_10_53_1","doi-asserted-by":"publisher","DOI":"10.1016\/B978-0-12-814976-8.00002-6"},{"key":"e_1_2_10_54_1","doi-asserted-by":"publisher","DOI":"10.1371\/journal.pone.0082971"},{"issue":"85","key":"e_1_2_10_55_1","first-page":"2825","article-title":"Scikit\u2010learn: machine learning in python","volume":"12","author":"Pedregosa F","year":"2011","journal-title":"Journal of Machine Learning Research"},{"key":"e_1_2_10_56_1","doi-asserted-by":"publisher","DOI":"10.1002\/lio2.259"},{"key":"e_1_2_10_57_1","unstructured":"Python Core Team. (2022)Python: a dynamic open source programming language. Python Software Foundation.https:\/\/www.python.org\/"},{"key":"e_1_2_10_58_1","unstructured":"R Core Team. (2022)R: a language and environment for statistical computing. R Foundation for Statistical Computing.https:\/\/www.r\u2010project.org\/"},{"key":"e_1_2_10_59_1","doi-asserted-by":"publisher","DOI":"10.1097\/01.mlg.0000179174.32345.41"},{"key":"e_1_2_10_60_1","doi-asserted-by":"publisher","DOI":"10.4236\/jsip.2014.51001"},{"key":"e_1_2_10_61_1","doi-asserted-by":"publisher","DOI":"10.1177\/0194599817751030"},{"key":"e_1_2_10_62_1","doi-asserted-by":"publisher","DOI":"10.1137\/S0036144598336745"},{"key":"e_1_2_10_63_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.protcy.2013.12.124"},{"key":"e_1_2_10_64_1","first-page":"44","volume-title":"Nature medicine (Vol. 25, Issue 1","author":"Topol E.J.","year":"2019"},{"key":"e_1_2_10_65_1","doi-asserted-by":"publisher","DOI":"10.1016\/J.JVOICE.2020.10.001"},{"key":"e_1_2_10_66_1","doi-asserted-by":"crossref","unstructured":"Wang Xia Dong Y. Hakkinen J.&Viikki O.(2000)Noise robust Chinese speech recognition using feature vector normalization and higher\u2010order cepstral coefficients. WCC 2000\u2014ICSP 2000 738\u2013741.https:\/\/doi.org\/10.1109\/ICOSP.2000.891617","DOI":"10.1109\/ICOSP.2000.891617"},{"key":"e_1_2_10_67_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.jvoice.2009.08.002"},{"key":"e_1_2_10_68_1","doi-asserted-by":"publisher","DOI":"10.1097\/AUD.0000000000000794"},{"key":"e_1_2_10_69_1","doi-asserted-by":"crossref","unstructured":"Xie Z. Gadepalli C. Jalalinajafabadi F. Cheetham B.M.G.&Homer J.J.(2018)Measurement of rater consistency and its application in voice quality assessments.Proceedings\u20142017 10th International Congress on Image and Signal Processing BioMedical Engineering and Informatics CISP\u2010BMEI 2017 2018 January 1\u20136.https:\/\/doi.org\/10.1109\/CISP\u2010BMEI.2017.8302302","DOI":"10.1109\/CISP-BMEI.2017.8302302"},{"key":"e_1_2_10_70_1","doi-asserted-by":"crossref","unstructured":"Yella S.H. Stolcke A.&Slaney M.(2014)Artificial neural network features for speaker diarization.2014 IEEE Workshop on Spoken Language Technology SLT 2014\u2014Proceedings 402\u2013406.https:\/\/doi.org\/10.1109\/SLT.2014.7078608","DOI":"10.1109\/SLT.2014.7078608"},{"key":"e_1_2_10_71_1","doi-asserted-by":"crossref","unstructured":"Yousef A.M. Deliyski D.D. Zacharias S.R.C.&Naghibolhosseini M.(2022)Detection of vocal fold image obstructions in high\u2010speed videoendoscopy during connected speech in adductor spasmodic dysphonia: a convolutional neural networks approach.Journal of Voice\u202f: Official Journal of the Voice Foundation https:\/\/doi.org\/10.1016\/J.JVOICE.2022.01.028","DOI":"10.1016\/j.jvoice.2022.01.028"},{"key":"e_1_2_10_72_1","doi-asserted-by":"publisher","DOI":"10.1121\/1.4964509"},{"key":"e_1_2_10_73_1","doi-asserted-by":"publisher","DOI":"10.1044\/1058\u20100360(2010\/09\u20100105"}],"container-title":["International Journal of Language &amp; Communication Disorders"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/1460-6984.12783","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/1460-6984.12783","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/1460-6984.12783","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2023,10,28]],"date-time":"2023-10-28T06:53:51Z","timestamp":1698476031000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/1460-6984.12783"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,9,18]]},"references-count":72,"journal-issue":{"issue":"2","published-print":{"date-parts":[[2023,3]]}},"alternative-id":["10.1111\/1460-6984.12783"],"URL":"https:\/\/doi.org\/10.1111\/1460-6984.12783","archive":["Portico"],"relation":{},"ISSN":["1368-2822","1460-6984"],"issn-type":[{"value":"1368-2822","type":"print"},{"value":"1460-6984","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,9,18]]},"assertion":[{"value":"2021-04-29","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-08-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2022-09-18","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}