{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T02:17:09Z","timestamp":1777861029753,"version":"3.51.4"},"reference-count":45,"publisher":"Wiley","issue":"5","license":[{"start":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T00:00:00Z","timestamp":1774310400000},"content-version":"vor","delay-in-days":0,"URL":"http:\/\/creativecommons.org\/licenses\/by\/4.0\/"},{"start":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T00:00:00Z","timestamp":1774310400000},"content-version":"tdm","delay-in-days":0,"URL":"http:\/\/doi.wiley.com\/10.1002\/tdm_license_1.1"}],"content-domain":{"domain":["onlinelibrary.wiley.com"],"crossmark-restriction":true},"short-container-title":["Expert Systems"],"published-print":{"date-parts":[[2026,5]]},"abstract":"<jats:title>ABSTRACT<\/jats:title>\n                  <jats:p>The continuous evolution of techniques for generating manipulated audio, known as voice deepfakes, and the widespread availability of tools that produce convincing forgeries have created an urgent need for reliable detection methods. This work considers the dimensionality of Mel\u2010Frequency Cepstral Coefficients (MFCCs) as a core design variable for practical, deployable systems. The aim is to identify the smallest number of coefficients that preserve detection performance across heterogeneous models while reducing computational cost, a critical factor for mobile and edge deployment. This study evaluates a hybrid setting on the ASVspoof 2019 Logical Access dataset, in which the same feature family serves as input to five traditional machine learning algorithms (Random Forest, k\u2010Nearest Neighbours, Linear Support Vector Classification, Extreme Gradient Boosting and Support Vector Machine with radial basis function kernel) and five deep learning models (Convolutional Neural Network, Recurrent Neural Network, Convolutional Recurrent Neural Network, Xception and ResNet). Results indicate that deep models reach near\u2010peak performance with a small number of coefficients, whereas classical methods require a larger number to achieve stable performance (except Linear Support Vector Classification, which consistently underperforms). Accordingly, 32 coefficients are considered an effective operating point for hybrid deployments. Overall, the results provide evidence to guide the selection of the number of MFCC coefficients in voice deepfake detection, aiming for efficient, reproducible and explainable systems.<\/jats:p>","DOI":"10.1111\/exsy.70245","type":"journal-article","created":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T08:38:28Z","timestamp":1774341508000},"update-policy":"https:\/\/doi.org\/10.1002\/crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["On the Optimal Selection of Mel\u2010Frequency Cepstral Coefficients for Voice Deepfake Detection"],"prefix":"10.1111","volume":"43","author":[{"ORCID":"https:\/\/orcid.org\/0009-0002-7106-6691","authenticated-orcid":false,"given":"Sergio A.","family":"Falc\u00f3n\u2010L\u00f3pez","sequence":"first","affiliation":[{"name":"Programa de Doctorado en Tecnolog\u00edas Industriales, Escuela Internacional de Doctorado UNED (EIDUNED) Universidad Nacional de Educacion a Distancia (UNED)  Madrid Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2779-4042","authenticated-orcid":false,"given":"Llanos","family":"Tobarra","sequence":"additional","affiliation":[{"name":"Control and Communication System Department, Computer Science Engineering Faculty Universidad Nacional de Educacion a Distancia (UNED)  Madrid Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5181-0199","authenticated-orcid":false,"given":"Antonio","family":"Robles\u2010G\u00f3mez","sequence":"additional","affiliation":[{"name":"Control and Communication System Department, Computer Science Engineering Faculty Universidad Nacional de Educacion a Distancia (UNED)  Madrid Spain"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4089-9538","authenticated-orcid":false,"given":"Rafael","family":"Pastor\u2010Vargas","sequence":"additional","affiliation":[{"name":"Control and Communication System Department, Computer Science Engineering Faculty Universidad Nacional de Educacion a Distancia (UNED)  Madrid Spain"}]}],"member":"311","published-online":{"date-parts":[[2026,3,24]]},"reference":[{"key":"e_1_2_9_2_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11227\u2010024\u201005960\u2010x"},{"key":"e_1_2_9_3_1","doi-asserted-by":"publisher","DOI":"10.3390\/forensicsci4030021"},{"key":"e_1_2_9_4_1","doi-asserted-by":"publisher","DOI":"10.3390\/computers13100256"},{"key":"e_1_2_9_5_1","doi-asserted-by":"crossref","unstructured":"AlSuwat M. S.Al\u2010Shareef M.Alghamdi andM.Almasre.2024.\u201cSpatial\u2010Temporal Self\u2010Supervised Learning for Audio Classification.\u201dSSRN.https:\/\/www.ssrn.com\/abstract=4882208.","DOI":"10.2139\/ssrn.4882208"},{"key":"e_1_2_9_6_1","doi-asserted-by":"publisher","DOI":"10.3389\/fdata.2024.1400024"},{"key":"e_1_2_9_7_1","doi-asserted-by":"publisher","DOI":"10.1145\/3543507.3583222"},{"key":"e_1_2_9_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/CAI64502.2025.00175"},{"key":"e_1_2_9_9_1","doi-asserted-by":"publisher","DOI":"10.1007\/s11042\u2010024\u201018217\u20109"},{"key":"e_1_2_9_10_1","doi-asserted-by":"publisher","DOI":"10.2478\/amns.2023.2.01376"},{"key":"e_1_2_9_11_1","doi-asserted-by":"crossref","unstructured":"Chen Y. J.Yi J.Xue et\u00a0al.2024.\u201cRawBMamba: End\u2010To\u2010End Bidirectional State Space Model for Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2406.06086.","DOI":"10.21437\/Interspeech.2024-698"},{"key":"e_1_2_9_12_1","doi-asserted-by":"publisher","DOI":"10.1111\/exsy.13322"},{"key":"e_1_2_9_13_1","unstructured":"Feng Y.2024.\u201cAudios Don't Lie: Multi\u2010Frequency Channel Attention Mechanism for Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2412.09467."},{"key":"e_1_2_9_14_1","unstructured":"Forum W. E.2025.\u201cGlobal Risks Report 2025.\u201dAccessed March 10 2026.https:\/\/www.weforum.org\/publications\/global\u2010risks\u2010report\u20102025\/."},{"key":"e_1_2_9_15_1","first-page":"12702","volume-title":"ICASSP 2024\u20132024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","author":"Guo Y.","year":"2024"},{"key":"e_1_2_9_16_1","unstructured":"Hsieh W. Z.Bi C.Jiang et\u00a0al.2024.\u201cA Comprehensive Guide to Explainable AI: From Classical Models to LLMs.\u201darXiv.https:\/\/arxiv.org\/abs\/2412.00800."},{"key":"e_1_2_9_17_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.asoc.2023.110124"},{"key":"e_1_2_9_18_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-97-0700-3_44"},{"key":"e_1_2_9_19_1","unstructured":"Kheir Y. E. Y.Samih S.Maharjan T.Polzehl andS.M\u00f6ller.2025.\u201cComprehensive Layer\u2010Wise Analysis of SSL Models for Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2502.03559."},{"key":"e_1_2_9_20_1","doi-asserted-by":"publisher","DOI":"10.1109\/WINCOM59760.2023.10323004"},{"key":"e_1_2_9_21_1","unstructured":"Le T. D. N. K. K.Teh andH. D.Tran.2024.\u201cContinuous Learning of Transformer\u2010Based Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2409.05924."},{"key":"e_1_2_9_22_1","doi-asserted-by":"publisher","DOI":"10.3390\/app13148488"},{"key":"e_1_2_9_23_1","doi-asserted-by":"crossref","unstructured":"Li X. K.Li Y.Zheng C.Yan X.Ji andW.Xu.2024.\u201cSafeEar: Content Privacy\u2010Preserving Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2409.09272.","DOI":"10.1145\/3658644.3670285"},{"key":"e_1_2_9_24_1","doi-asserted-by":"crossref","unstructured":"Liu R. J.Zhang G.Gao andH.Li.2023.\u201cBetray Oneself: A Novel Audio DeepFake Detection Model via Mono\u2010To\u2010Stereo Conversion.\u201darXiv.https:\/\/arxiv.org\/abs\/2305.16353.","DOI":"10.21437\/Interspeech.2023-2335"},{"key":"e_1_2_9_25_1","doi-asserted-by":"crossref","unstructured":"Ma Y. Z.Ren andS.Xu.2021.\u201cRW\u2010Resnet: A Novel Speech Anti\u2010Spoofing Model Using Raw Waveform.\u201darXiv.https:\/\/arxiv.org\/abs\/2108.05684.","DOI":"10.21437\/Interspeech.2021-438"},{"key":"e_1_2_9_26_1","unstructured":"Magazine C.2025.\u201cHuman Error and the Financial Impact of Cybercrime.\u201dAccessed March 10 2026.https:\/\/cybersecurityventures.com\/human\u2010error\u2010and\u2010the\u2010financial\u2010impact\u2010of\u2010cybercrime\/."},{"key":"e_1_2_9_27_1","doi-asserted-by":"publisher","DOI":"10.1109\/APSIPAASC63619.2025.10848893"},{"key":"e_1_2_9_28_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.procs.2023.01.283"},{"key":"e_1_2_9_29_1","doi-asserted-by":"publisher","DOI":"10.25195\/ijci.v50i2.506"},{"key":"e_1_2_9_30_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSTSP.2017.2684705"},{"key":"e_1_2_9_31_1","unstructured":"Pham L. P.Lam T.Nguyen H.Nguyen andA.Schindler.2024.\u201cDeepfake Audio Detection Using Spectrogram\u2010Based Feature and Ensemble of Deep Learning Models.\u201darXiv.https:\/\/arxiv.org\/abs\/2407.01777."},{"key":"e_1_2_9_32_1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-77299-3_19"},{"key":"e_1_2_9_33_1","doi-asserted-by":"publisher","DOI":"10.20998\/2522\u20109052.2024.2.10"},{"key":"e_1_2_9_34_1","doi-asserted-by":"publisher","DOI":"10.1109\/IJCB62174.2024.10744454"},{"key":"e_1_2_9_35_1","doi-asserted-by":"crossref","unstructured":"Rimon I. O.Gal andH.Permuter.2025.\u201cUnmasking Deepfakes: Leveraging Augmentations and Features Variability for Deepfake Speech Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2501.05545.","DOI":"10.2139\/ssrn.5184289"},{"key":"e_1_2_9_36_1","doi-asserted-by":"crossref","unstructured":"Saha S. M.Sahidullah andS.Das.2024.\u201cExploring Green AI for Audio Deepfake Detection.\u201darXiv.https:\/\/arxiv.org\/abs\/2403.14290.","DOI":"10.23919\/EUSIPCO63174.2024.10715424"},{"key":"e_1_2_9_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/3643491.3660289"},{"key":"e_1_2_9_38_1","unstructured":"Shin H.\u2010S. J.Heo J.\u2010H.Kim C.\u2010Y.Lim W.Kim andH.\u2010J.Yu.2023.\u201cHM\u2010Conformer: A Conformer\u2010Based Audio Deepfake Detection System With Hierarchical Pooling and Multi\u2010Level Classification Token Aggregation Methods.\u201darXiv.https:\/\/arxiv.org\/abs\/2309.08208."},{"key":"e_1_2_9_39_1","doi-asserted-by":"publisher","DOI":"10.52783\/jes.1829"},{"key":"e_1_2_9_40_1","doi-asserted-by":"publisher","DOI":"10.12928\/TELKOMNIKA.v21i1.24259"},{"key":"e_1_2_9_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2024.3521026"},{"key":"e_1_2_9_42_1","doi-asserted-by":"publisher","DOI":"10.1145\/3552466.3556530"},{"key":"e_1_2_9_43_1","unstructured":"Wu H. J.Chen R.Du et\u00a0al.2024.\u201cCLAD: Robust Audio Deepfake Detection Against Manipulation Attacks With Contrastive Learning.\u201darXiv.https:\/\/arxiv.org\/abs\/2404.15854."},{"key":"e_1_2_9_44_1","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2023-1383"},{"key":"e_1_2_9_45_1","doi-asserted-by":"publisher","DOI":"10.1145\/3552466.3556526"},{"key":"e_1_2_9_46_1","unstructured":"Yi J. C. Y.Zhang J.Tao et\u00a0al.2024.\u201cADD 2023: Towards Audio Deepfake Detection and Analysis in the Wild.\u201darXiv.https:\/\/arxiv.org\/abs\/2408.04967."}],"container-title":["Expert Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70245","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/full-xml\/10.1111\/exsy.70245","content-type":"application\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/pdf\/10.1111\/exsy.70245","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,30]],"date-time":"2026-04-30T02:44:12Z","timestamp":1777517052000},"score":1,"resource":{"primary":{"URL":"https:\/\/onlinelibrary.wiley.com\/doi\/10.1111\/exsy.70245"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,24]]},"references-count":45,"journal-issue":{"issue":"5","published-print":{"date-parts":[[2026,5]]}},"alternative-id":["10.1111\/exsy.70245"],"URL":"https:\/\/doi.org\/10.1111\/exsy.70245","archive":["Portico"],"relation":{},"ISSN":["0266-4720","1468-0394"],"issn-type":[{"value":"0266-4720","type":"print"},{"value":"1468-0394","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,24]]},"assertion":[{"value":"2025-10-30","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-05","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-03-24","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}],"article-number":"e70245"}}