A Review of Modern Audio Deepfake Detection Methods: Challenges and Future Directions
Open accessThe DOI matches its Crossref record; 56 of the 56 references it cites were found in a registry or an index.
Abstract
A number of AI-generated tools are used today to clone human voices, leading to a new technology known as Audio Deepfakes (ADs). Despite being introduced to enhance human lives as audiobooks, ADs have been used to disrupt public safety. ADs have thus recently come to the attention of researchers, with Machine Learning (ML) and Deep Learning (DL) methods being developed to detect them. In this article, a review of existing AD detection methods was conducted, along with a comparative description of the available faked audio datasets. The article introduces types of AD attacks and then outlines and analyzes the detection methods and datasets for imitation- and synthetic-based Deepfakes. To the best of the authors’ knowledge, this is the first review targeting imitated and synthetically generated audio detection methods. The similarities and differences of AD detection methods are summarized by providing a quantitative comparison that finds that the method type affects the performance more than the audio features themselves, in which a substantial tradeoff between the accuracy and scalability exists. Moreover, at the end of this article, the potential research directions and challenges of Deepfake detection methods are discussed to discover that, even though AD detection is an active area of research, further research is still needed to address the existing gaps. This article can be a starting point for researchers to understand the current state of the AD literature and investigate more robust detection models that can detect fakeness even if the target audio contains accented voices or real-world noises.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Traditional deep learning approaches assume that training and testing data share a static distribution, which is unrealistic in scenarios where new spoofing attacks emerge regularly [59].”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [59]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Traditional deep learning approaches assume that training and testing data share a static distribution, which is unrealistic in scenarios where new spoofing attacks emerge regularly [59].”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [59]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Deep learning models face significant challenges in handling dynamically evolving data distributions, particularly in areas such as audio deepfake detection.”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [59]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Today, deepfakes can be generated in real-time, enabling attackers to impersonate people over audio and video calls (Almutairi & Elgibreen, 2022).”
From Deepfake forensics: a survey of digital forensic methods for multimodal deepfake identification on social media (2024) · cites this paper as Almutairi & Elgibreen, 2022
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“The need for more robust detection models to detect fakeness even in the presence of accented voices or real-world noises is highlighted.”
From Deepfake forensics: a survey of digital forensic methods for multimodal deepfake identification on social media (2024) · cites this paper as Almutairi & Elgibreen, 2022
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Natural TTS Synthesis by Conditioning Wavenet on MEL Spectrogram Predictions
Shen, J., Pang, R., Weiss, R. J., Schuster, M., Jaitly, N., Yang, Z., Chen, Z., Zhang, Y., Wang, Y., Skerrv-Ryan, R., et al. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461368 · Cited by 2,791
Self-supervised Learning: Generative or Contrastive
Liu, X., Zhang, F., Hou, Z., Mian, L., Wang, Z., Zhang, J., & Tang, J. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3090866 · Cited by 1,087 · cited 1×
“Following this, the available faked audio datasets were summarized, and the discussed methods were also compared.”
Synthesizing Obama: learning lip sync from audio
Suwajanakorn, S., Seitz, S. M., & Kemelmacher-Shlizerman, I. (2017) · ACM Transactions on Graphics · doi:10.1145/3072959.3073640 · Cited by 1,057
ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
Todisco, M., Wang, X., Vestman, V., Sahidullah, M., Delgado, H., Nautsch, A., Yamagishi, J., Evans, N., Kinnunen, T., & Lee, K. A. (2019) · arXiv · doi:10.48550/arxiv.1904.05441 · Cited by 688
ASVspoof 2019: Future Horizons in Spoofed and Fake Audio Detection
Todisco, M., Wang, X., Vestman, V., Sahidullah, M., Delgado, H., Nautsch, A., Yamagishi, J., Evans, N., Kinnunen, T. H., & Lee, K. A. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2249 · Cited by 687
FastSpeech 2: Fast and High-Quality End-to-End Text to Speech
Ren, Y., Hu, C., Tan, X., Qin, T., Zhao, S., Zhao, Z., & Liu, T.-Y. (2020) · arXiv · doi:10.48550/arxiv.2006.04558 · Cited by 580
ASVspoof 2015: the first automatic speaker verification spoofing and countermeasures challenge
Wu, Z., Kinnunen, T., Evans, N., Yamagishi, J., Hanilçi, C., Sahidullah, M., & Sizov, A. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-462 · Cited by 484
Deep voice 3: scaling text-to-speech with convolutional sequence learning
Ping, W., Peng, K., Gibiansky, A., Arik, S. O., Kannan, A., Narang, S., Raiman, J., & Miller, J. (2017) · arXiv · doi:10.48550/arxiv.1710.07654 · Cited by 313
Deepfake Detection: Current Challenges and Next Steps
Lyu, S. (2020) · 2020 IEEE International Conference on Multimedia & Expo Workshops (ICMEW) · doi:10.1109/icmew46912.2020.9105991 · Cited by 234
Recurrent convolutional structures for audio spoof and video deepfake detection
Chintha, A., Thai, B., Sohrawardi, S. J., Bhatt, K., Hickerson, A., Wright, M., & Ptucha, R. (2020) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2020.2999185 · Cited by 226 · cited 1×
“They used an image-based approach, which required a special transformation of the data, and the authors found that the model proposed failed to correctly classify new audio signals indicating that the model is not general enough.”
Research works printed but not linked: 3
- Khanjani, Z., Watson, G., and Janeja, V.P. (2021). How deep are the fakes? Focusing on audio deepfake: A survey. arXiv.
- Almeman, K., and Lee, M. (2013, January 16–19). A comparison of arabic speech recognition for multi-dialect vs. specific dialects. Proceedings of the Seventh International Conference on Speech Technology and Human-Computer Dialogue (SpeD 2013), Cluj-Napoca, Romania.
- Jain, 2022, Review paper on noise cancellation using adaptive filters, Int. J. Eng. Res. Technol., 11, 241
Cited by
Items 201–228 of 228
Feature Extraction from AI generated songs
Adda, M., Scheidt, N., GEGOV, A. E., Ahmad, F., Sivaraman, P. R., & Ichtev, A. (2026) · doi:10.1109/icarai70085.2026.11635738
“Early detection systems for AI-generated voices have demonstrated partial success but often struggle to generalise across diverse datasets or maintain consistent accuracy [3].”
An Efficient Deepfake Arabic Voice Detection System Based on a Deep Learning Approach
Almarashi, R., & Kim, J. (2026) · Studies in big data · doi:10.1007/978-3-032-29599-6_5
Sofyan, M., Suryanata, M. G., & Wahyuni, M. S. (2026) · Jurnal Sistem Informasi Triguna Dharma (JURSI TGD) · doi:10.53513/jursi.v5i3.10434
“Berdasarkan hasil review yang dilakukan Almutairi dan Elgibren [23], penelitian yang menggunakan metode deep learning khususnya CNN menjadi pilihan yang terbaik serta memiliki kinerja dan stabilitas yang mumpuni dibandingkan dengan berbasis machine learning , seperti Support Vector Machine (SVM).”
Deep Learning based DeepFake Image Detection for Enhancing Social Media Security
Patil, P., & Chopde, N. (2026) · International Conference on Sustainable Computing and Smart Systems (ICSCSS) · doi:10.1109/icscss69635.2026.11645824
“Therefore, there is a need for a deepfake detection system that automatically distinguishes original content from deepfakes to prevent the misuse of original content on social media platforms [4], [5].”
Deepfakes and identity theft in the age of AI
Nurguzhina, A., Sarsenbi, Z., Umbetalina, S., & Kaldybek, A. (2026) · International Conference on Cybersecurity, Digital Forensics, and AI Applications (ICCSDFAI) · doi:10.1109/iccsdfai70505.2026.11647888
“Detection pipelines can be highly effective under benchmark conditions, yet their performance degrades when the content is compressed, translated across platforms, or partially manipulated in only one channel [18], [20], [21], [31].”
PAttSBiL: an efficient deep fake video detection using integrated deep learning methodologies
Agrawal, D. R., & Haneef, F. (2026) · Multimedia Tools and Applications · doi:10.1007/s11042-026-21854-x
“Combining deep learning with false, deepfake films use techniques such total face syn - thesis, face editing, face reenactment, and face replacement to give the impression that one person's face is superimposed onto another's [ 7].”
Pethkar, S., Shelkar, S., Pandey, S., & Bhagat, D. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-2600-1990-0.ch012
Audio Deepfake Detection: A Comprehensive Survey of Generation, Datasets, and Detection Strategies
Abbas, S. S., Al-Obaidi, S., & Al-Sultan, A. Y. (2026) · International Journal of Safety and Security Engineering · doi:10.18280/ijsse.160718
Saputra, D. A., Saputro, N. D., & Renaldy, R. (2026) · Brilliance Research of Artificial Intelligence · doi:10.47709/brilliance.v6i3.9372
“Each segment becomes a mel-spectrogram through STFT and mel-scale mapping, consistent with Almutairi & Elgibreen (2022), Yang et al. (2022), Zaman, Sah, Direkoglu, & Unoki (2023), and Gong, Chung, & Glass (2021a).”
Deepfake recognition via cascaded neuro-fuzzy learning and parallel convolutional harmonic networks
Raj, S., Agrawal, A. P., Meshram, P., & Mahajan, G. (2026) · Discover Applied Sciences · doi:10.1007/s42452-026-09172-1
“Therefore, detecting deepfake image content plays a vital role in identifying these manipulations [9].”
Spanish Audio DeepFake Detection
Torres, D. B., Vargas, J. I. D. L. R., Sánchez, A. B., Moreno, G., & González, E. (2026) · Communications in computer and information science · doi:10.1007/978-3-032-37438-7_35
Deepfake Threats in Business Communications: Detection, Governance, and Trust Management
Elsayed, S. M. (2026) · World Research of Business Administration Journal · doi:10.56830/wrba07202601
“(Almutairi & Elgibreen, 2022).”
Advancing Multi-Speaker Audio Deepfake Detection through Chunk-Based Evidence Aggregation
Ahmed, A. J., & Janeja, V. P. (2026) · IEEE International Conference on Intelligence and Security Informatics (ISI) · doi:10.1109/isi69914.2026.11685842
“…have explored both handcrafted features such as Mel-frequency cepstral coefficients (MFCCs) [16] and Linear Frequency Cepstral Coefficients (LFCC) [17] with classical classifiers, and deep learning approaches [18], [19] such as RawNet2 [20], which learns representations directly from raw waveforms.”
Study of Deepfake Audio Detection in Telephone Speech with Pretrained Self-supervised Models
Sielecki, S., Kisielewicz, M., & Kędziora, M. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-36868-3_24
Moallim, M., Alhaj, T. A., Elhaj, F. A., & Darwish, T. (2026) · IEEE Access · doi:10.1109/access.2026.3734022
“Several studies have reviewed existing audio deepfake detection corpora, including those by Almutairi et al. [25], Yi et al. [2], and Zhang et al. [1].”
Innovative Technologies for Creating Multilingual Audio content in the Publishing Industry
Kalmykov, A. (2023) · Law Business and Sustainability Herald · Cited by 1
Analysing the landscape of Deep Fake Detection: A Survey
Vyas, K., Pareek, P., Jayaswal, R., & Patil, S. (2024) · International Journal of Intelligent Systems and Applications in Engineering · Cited by 12
An Edge Device for Real-Time Fake and Tampered Audio Detection
Bobate, R. V., Arora, K., Tirale, H., Admane, A., & Jambhulkar, F. (2026) · INTERANTIONAL JOURNAL OF SCIENTIFIC RESEARCH IN ENGINEERING AND MANAGEMENT · doi:10.55041/ijsrem67662
Li, Y., Fang, Y., Xue, J., Milling, M., & Schuller, B. W. (2026) · Frontiers in Artificial Intelligence · doi:10.3389/frai.2026.1928592
IndianDeepGuard: Enhancing Deepfake Detection through Cultural and Multi-Modal Context
Monpara, K., Savla, J., Sanura, M., Shah, M., Tawde, P., & Agarwal, N. (2026) · Advances in intelligent systems research/Advances in Intelligent Systems Research · doi:10.2991/978-94-6239-791-0_8
An overview of Challenges in Deepfake technology-detection and Mitigation
Shetty, A., & Hariharasitaraman, S. (2026) · International Conference on Advancements in Smart, Secure and Intelligent Computing (ASSIC) · doi:10.1109/assic70037.2026.11716950
“Explainable detection is therefore preferred in many cases, since the result can be connected to visible regions or meaningful features instead of being reported only as a “black - box” label [1 4].”
Convolutional Neural Network-Based Detection and Identification of Deepfake Audio
Okunade, O. A. (2022) · International Journal of Science for Global Sustainability
Olugbemi-Bello, M. A., Chiagozie, U. V., Yesuf, F., & Sunmonu, B. J. (2026) · Algora · doi:10.63084/algora.v3i1.119
“Almutairi and Elgibreen (2022) report that the type of method affects detection more than the acoustic features used, and they note difficulties with accented speech and environmental noise.”
Audio Deepfake: Literature Review and Detection Limitations
Mahmoud, B., & Bouzidi, D. (2026) · Lecture notes on data engineering and communications technologies · doi:10.1007/978-3-032-18927-1_15
Zhang, S., Zheng, P., Yang, Y., Jiang, Q., & Long, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-42929-2_16
Сравнительный анализ XAI-подходов для обнаружения синтетической речи
Ким, К. К. (2026) · Problems of Computational and Applied Mathematics · doi:10.71310/pcam.5_76.2026.06
Acoustic Microcues: Breath and Phoneme-Level Signals for Detecting Audio Deepfakes
Han, W., Palasamudram, A., Fullwood, J., & Monrose, F. (2026) · doi:10.1145/3840473.3840514
Bhandari, P., Neupane, P., Marasini, A., & Sharma, B. (2026) · Morgan Journal of Interdisciplinary Research Studies · doi:10.3126/mjirs.v2i1.101306
Metadata
- Authors
- Almutairi, Z., & Elgibreen, H.
- Affiliations
- King Saud University (SA)
- Corresponding author
- Almutairi, Zaynab
- Venue
- Algorithms
- Publisher
- MDPI AG
- Published
- 2022-05-04
- Volume
- 15
- Issue
- 5
- Pages
- 155
- ISSN
- 1999-4893 (ISSN-L)
- OpenAlex
- W4225420087
- Cited by
- 187
- References
- 46
- Citations by year
- 2026: 46 · 2025: 72 · 2024: 47 · 2023: 21 · 2022: 1
- FWCI
- 11.60
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Algorithms (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- audio deepfake detection, voice cloning, AI-generated audio, synthetic speech, audio forensics
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.