CoChat
Sign Up
Cited by 56Speech Recognition and SynthesisDigital Media Forensic DetectionMusic and Audio Processing

Review of audio deepfake detection techniques: Issues and prospects

Open access
Dixit, Abhishek, Kaur, Nirmal, Kingra, Staffy·2023·Expert Systems
DOI 10.1111/exsy.13322
Open this paper in CoChat — free accountOpen PDF

Abstract

In the past years, multimedia content has improved in realism and plausibility owing to the development of deep learning techniques, particularly the generative adversarial networks and variational auto‐encoders. Though digital content, especially digital movies shot from a certain viewpoint gives a true representation of reality, yet the ubiquitous usage of content manipulation techniques casts doubt on its veracity. Deepfaking an AI based tampering technique, is able to map facial and acoustic features of a source person onto the target with an intention to make target say or enact the things that has not happened in real. Numerous approaches have been proposed in the literature for detection of image and video deepfakes. With technological advancement, researchers have also started to examine audio deepfakes and ways to detect them. As there is currently no comprehensive overview of audio deepfake generation and detection techniques, this paper aims to provide a survey of the relevant literature in this area. This survey paper intends to help research fraternity about the available audio generation and detection approaches for design of reliable detection models in future to classify fake and real audios.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1111/exsy.13322

Cites

5 of 5 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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?

  • Traditional detection methods rely on classical machine learning models and often struggle with generalization and scalability when confronted with the latest GANs and advanced deep learning architectures.

    “Traditional detection methods [3], [4] rely on classical machine learning models; however, these approaches often struggle with generalization and scalability when confronted with the latest generative adversarial networks (GANs) and advanced deep learning architectures that are used to create deepfakes [5].”

    From Quantum-Trained Convolutional Neural Network for Deepfake Audio Detection (2025) · cites this paper as [4]

    Section 4.3 says SVM alone was not very scalable because it requires manual feature extraction, and Section 5 says detection methods need better generalizability on realistic deepfake audio. The paper does not specifically establish that traditional methods struggle with both problems when confronted with the latest GANs and advanced deep-learning generators. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 52

  1. Albadawy, E. A., Lyu, S., & Farid, H. Detecting AI-synthesized speech using bispectral analysis. Almeman, K. (2013). A comparison of Arabic speech recognition for multi-dialect vs. specific dialects. Retrieved January 4, 2023, from https://www. researchgate.net/profile/Khalid-Almeman/publication/308886826_A_Comparison_of_Arabic_Speech_Recognition_for_Multi-Dialect_vs_Specific_Dialects/ links/57f4a5e908ae886b897de423/A-Comparison-of-Arabic-Speech-Recognition-for-Multi-Dialect-vs-Specific-Dialects.pdf
  2. Almutairi, Z., & Elgibreen, H. (2022). A review of modern audio deepfake detection methods: Challenges and future directions. Algorithms, 15(5), 1-2. https://doi.org/10.3390/a15050155 · DOI · Paper page
  3. Balamurali, B., Lin, K., & Lui, S. Toward robust audio spoofing detection: A detailed comparison of traditional and learned features. Retrieved January 4, 2023, from https://ieeexplore.ieee.org/abstract/document/8740863/
  4. Ballesteros, D., & Rodriguez-Ortega, Y. (2021). Deep4SNet: Dekingraep learning for fake speech classification. Retrieved January 4, 2023, from https:// www.sciencedirect.com/science/article/pii/S0957417421008770?casa_token=AnsH7MnoSZ0AAAAA:JwTe9BOGj0wck0YM0KePOPSaP3t7Pag0pjW bNs0tjyhyv4OTuqDizgYbd9myzYRPLwxRMbow4RM
  5. Ballesteros, D. M., Rodriguez, Y., & Renza, D. (2020). A dataset of histograms of original and fake voice recordings (H-Voice). Data in Brief, 29, 105331. https://doi.org/10.1016/j.dib.2020.105331 · DOI
  6. Borrelli, C., Bestagini, P., Antonacci, F., Sarti, A., & Tubaro, S. (2021). Synthetic speech detection through short-term and long-term prediction traces. EURASIP Journal on Information Security, 2021(1), 2-3. https://doi.org/10.1186/S13635-021-00116-3 · DOI · Paper page
  7. Dolhansky, B., Howes, R., Pflaum, B., Baram, N., & Ferrer, C. C. The Deepfake Detection Challenge (DFDC) preview dataset. Retrieved from http://arxiv. org/abs/1910.08854
  8. Frank, J., & Schönherr, L. WaveFake: A data set to facilitate audio deepfake detection. Retrieved November 2021, from http://arxiv.org/abs/2111.02813 · arXiv
  9. Garrido, P., Valgaerts, L., Sarmadi, H., Steiner, I., Varanasi, K., Pérez, P., & Theobalt, C. (2015). VDub: Modifying face video of actors for plausible visual alignment to a dubbed audio track. Computer Graphics Forum, 34(2), 193-204. https://doi.org/10.1111/CGF.12552 · DOI
  10. GitHub. (2023). GitHub-resemble-ai/Resemblyzer: A python package to analyze and compare voices with deep learning. Retrieved January 4, 2023, from https://github.com/resemble-ai/Resemblyzer Contributing data to Deepfake detection research-Google AI Blog. Retrieved January 4, 2023, from https://ai.googleblog.com/2019/09/contributing- data-to-deepfake-detection.html IEEE Signal Processing Society, Yang, X., Li, Y., & Lyu, S. (2018). IEEE International Conference on Acoustics, Speech and Signal Processing : Proceedings : April 15-20, 2018. Calgary Telus Convention Center.
  11. Ito, K., & Johnson, L. (2023). @misc{ljspeech17}-Google Scholar. Retrieved January 4, 2023, from https://scholar.google.com/scholar?hl=en&as_sdt=0% 2C5&q=%40misc%7Bljspeech17%2C+author+%3D+%7BKeith+Ito+and+Linda+Johnson%7D%2C+title+%3D+%7BThe+LJ+Speech+Dataset %7D%2C+howpublished+%3D+%7B%5Curl%7Bhttps%3A%2F%2Fkeithito.com%2FLJ-Speech-Dataset%2F%7D%7D%2C+year+%3D+2017+% 7D&btnG=
  12. Karras, T., Aila, T., Laine, S., & Lehtinen, J. (2017). Progressive growing of GANs for improved quality, stability, and variation. Retrieved from http://arxiv. org/abs/1710.10196
  13. Karras, T., Laine, S., & Aila, T. (2019). A style-based generator architecture for generative adversarial networks. Retrieved January 4, 2023, from http:// openaccess.thecvf.com/content_CVPR_2019/html/Karras_A_Style-Based_Generator_Architecture_for_Generative_Adversarial_Networks_CVPR_2019_ paper.html
  14. Khalid, H., Kim, M., & Tariq, S. (2021). Evaluation of an audio-video multimodal deepfake dataset using unimodal and multimodal detectors (pp. 7-15). https://doi.org/10.1145/3476099.3484315 · DOI · Paper page
  15. Khalid, H., Tariq, S., Kim, M., & Woo, S. S. FakeAVCeleb: A novel audio-video multimodal Deepfake dataset. Retrieved from http://arxiv.org/abs/2108. 05080
  16. Kingma, D. P., & Welling, M. (2014). Auto-encoding variational bayes [Conference presentation]. Second International Conference on Learning Representa- tions, ICLR 2014-Conference Track Proceedings.
  17. Kingra, S., Aggarwal, N., & Kaur, N. (2022a). LBPNet: Exploiting texture descriptor for deepfake detection. Forensic Science International: Digital Investigation, 42, 301452. https://doi.org/10.1016/j.fsidi.2022.301452 · DOI
  18. Kingra, S., Aggarwal, N., & Kaur, N. (2022b). Emergence of deepfakes and video tampering detection approaches: A survey. Multimedia Tools and Applica- tions, 82(7), 10165-10209. https://doi.org/10.1007/s11042-022-13100-x · DOI · Paper page
  19. Lai, C. I., Chen, N., Villalba, J., & Dehak, N. (2019). ASSERT: Anti-spoofing with squeeze-excitation and residual networks. Proceedings of the Annual Confer- ence of the International Speech Communication Association INTERSPEECH, 2019, 1013-1017. https://doi.org/10.21437/Interspeech.2019-1794 · DOI
  20. Lataifeh, M., Elnagar, A. M., Shahin, I., Nassif, A. B., & Elnagar, A. (2020). Arabic audio clips: Identification and discrimination of authentic cantillations from imitations. Neurocomputing, 418, 162-177. https://doi.org/10.1016/j.neucom.2020.07.099 · DOI
  21. Liu, M., & Tuzel, O. (2023). Coupled generative adversarial networks. Retrieved January 4, 2023, from https://proceedings.neurips.cc/paper/6544- coupled-generative-adversarial-networks
  22. Liu, T., Yan, D., Wang, R., & Yan, N. (2021). Identification of fake stereo audio using SVM and CNN. Retrieved January 4, 2023, from https://www.mdpi. com/2078-2489/12/7/263
  23. Masood, M., Nawaz, M., Malik, K. M., Javed, A., Irtaza, A., & Malik, H. (2022). Deepfakes generation and detection: State-of-the-art, open challenges, coun- termeasures, and way forward. Applied Intelligence, 53, 3974-4026. https://doi.org/10.1007/S10489-022-03766-Z · DOI · Paper page
  24. Ning, Y., He, S., Wu, Z., Xing, C., & Zhang, L. J. (2019). Review of deep learning based speech synthesis. Applied Sciences (Switzerland), 9(19), 9-12. https:// doi.org/10.3390/app9194050 · DOI
  25. Pianese, A., & Cozzolino, D. (2022). Deepfake audio detection by speaker verification. Retrieved January 4, 2023, from https://ieeexplore.ieee.org/ abstract/document/9975428/?casa_token=UDQivgjF3iQAAAAA:zPWbdqYPYuhW2W6Abx_Z13ZmRqIp2ZS5QxRg24y2GvTJ0iluTSUX6oKp123IeSg Q3bguqFolTj0
  26. Radford, A., Metz, L., & Chintala, S. (2016). Unsupervised representation learning with deep convolutional generative adversarial networks [Conference presen- tation]. Fourth International Conference on Learning Representations, ICLR 2016-Conference Track Proceedings.
  27. Reimao, R., & Tzerpos, V. (2019). FoR: A dataset for synthetic speech detection [Conference presentation]. Tenth International Conference on Speech Tech- nology and Human-Computer Dialogue, SpeD 2019. https://doi.org/10.1109/SPED.2019.8906599 · DOI · Paper page
  28. Ren, Y., Hu, C., Tan, X., Qin, T., Zhao, S., Zhao, Z., & Liu, T.-Y. (2020). FastSpeech 2: Fast and high-quality end-to-end text to speech. Retrieved from http:// arxiv.org/abs/2006.04558 · arXiv
  29. Rodríguez-Ortega, Y., Ballesteros, D. M., & Renza, D. (2020). A machine learning model to detect fake voice. Communications in Computer and Information Science, 1277, 3-13. https://doi.org/10.1007/978-3-030-61702-8_1/COVER · DOI
  30. Rossler, A., Cozzolino, D., Verdoliva, L., Riess, C., Thies, J., & Niessner, M. (2019). FaceForensics++: Learning to detect manipulated facial images. Proceed- ings of the IEEE International Conference on Computer Vision, 2019, 1-11. https://doi.org/10.1109/ICCV.2019.00009 · DOI · Paper page
  31. Saranya, M. S., Padmanabhan, R., & Murthy, H. A. (2018). Replay attack detection in speaker verification using non-voiced segments and decision level fea- ture switching. Twelfth International Conference on Signal Processing and Communications: 16th-19th July 2018, Indian Institute of Science (pp. 332-336).
  32. Sarfjoo, S. S., Wang, X., Henter, G. E., Lorenzo-Trueba, J., Takaki, S., & Yamagishi, J. (2019). Transformation of low-quality device-recorded speech to high- quality speech using improved SEGAN model. https://doi.org/10.7488/ds/1994 · DOI
  33. Sercan, S., Chrzanowski, M., Coates, A., Diamos, G., Gibiansky, A., Kang, Y., Li, X., Miller, J., Ng, A., Raiman, J., Sengupta, S., & Shoeybi, M. (2017). Deep voice: Real-time neural text-to-speech. PMLR (pp. 195-204). Retrieved January 4, 2023, from https://proceedings.mlr.press/v70/arik17a.html
  34. Shan, M., & Tsai, T. (2020). A cross-verification approach for protecting world leaders from fake and tampered audio. Retrieved January 4, 2023, from http://arxiv.org/abs/2010.12173 · arXiv
  35. Singh, A. (2020). Detection of ai-synthesized speech using cepstral and bispectral statistics. Retrieved January 4, 2023, from https://ieeexplore.ieee.org/ abstract/document/9565457/?casa_token=TE7abzxYC_8AAAAA:zczY63UDhdmhDClCekM6_6m3I0WBowFbZ15PcO5RORZFipg6NPhEm5O8HRSrS VibZ3jXqWsFj3g
  36. Sisman, B., Yamagishi, J., King, S., & Li, H. (2021). An overview of voice conversion and its challenges: From statistical modeling to deep learning. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 29, 132-157. https://doi.org/10.1109/TASLP.2020.3038524 · DOI
  37. Subramani, N. Learning efficient representations for fake speech detection. Retrieved January 4, 2023, from https://ojs.aaai.org/index.php/AAAI/article/ view/6044
  38. Suwajanakorn, S., Seitz, S. M., & Kemelmacher-Shlizerman, I. (2017). Synthesizing Obama: Learning lip sync from audio. ACM Transactions on Graphics, 36(4), 1-14. https://doi.org/10.1145/3072959.3073640 · DOI · Paper page
  39. Tan, X., Qin, T., Soong, F., & Liu, T.-Y. (2021). A survey on neural speech synthesis. Retrieved from http://arxiv.org/abs/2106.15561 · arXiv
  40. van den Oord, A., Li, Y., Babuschkin, I., Simonyan, K., Vinyals, O., Kavukcuoglu, K., van den Driessche, G., Lockhart, E., Cobo, L. C., Stimberg, F., Casagrande, N., Grewe, D., Noury, S., Dieleman, S., Elsen, E., Kalchbrenner, N., Zen, H., Graves, A., King, H., … Hassabis, D. (2018). Parallel WaveNet: Fast high-fidelity speech synthesis PMLR (pp. 3918-3926). Retrieved January 4, 2023, from https://proceedings.mlr.press/v80/oord18a.html
  41. Wang, R., Juefei-Xu, F., Huang, Y., Guo, Q., Xie, X., Ma, L., & Liu, Y. (2020). DeepSonar: Towards effective and robust detection of AI-synthesized fake voices [Conference presentation]. MM 2020-Proceedings of the 28th ACM International Conference on Multimedia (pp. 1207-1216). https://doi.org/10. 1145/3394171.3413716 · DOI · Paper page
  42. Wang, R., Ding, Y., Li, L., & Fan, C. (2020). IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2020). IEEE.
  43. Wang, Y., Skerry-Ryan, R. J., Stanton, D., Wu, Y., Weiss, R. J., Jaitly, N., Yang, Z., Xiao, Y., Chen, Z., Bengio, S., Le, Q., Agiomyrgiannakis, Y., Clark, R., & Saurous, R. A. (2017). Tacotron: Towards end-to-end speech synthesis. Retrieved from http://arxiv.org/abs/1703.10135 · arXiv
  44. Wijethunga, R. (2020). Deepfake audio detection: A deep learning based solution for group conversations. Retrieved January 4, 2023, from https:// ieeexplore.ieee.org/abstract/document/9357161/?casa_token=KO72Bp_vhnwAAAAA:dFgt6jgqerwao_qnB1f7hRIVBk4RclcLEM12WVq7RZ2fJkOLA9 hziZzhfw74toudgyefCCCkEdg
  45. Witkowski, M., Kacprzak, S., Kowalczyk, K., Zelasko, P., & Gałka, J. (2017). Audio replay attack detection using high-frequency features. https://doi.org/10. 21437/Interspeech.2017-776 · DOI
  46. Xie, Y., Shi, C., Li, Z., Liu, J., Chen, Y., & Yuan, B. (2020). Real-time, universal, and robust adversarial attacks against speaker recognition systems. In ICASSP 2020 -2020 IEEE international conference on acoustics, speech and signal processing (ICASSP) (pp. 1738-1742). IEEE. https://doi.org/10.1109/ ICASSP40776.2020.9053747 · DOI
  47. Yu, H., Tan, Z., & Ma, Z. (2017). Spoofing detection in automatic speaker verification systems using DNN classifiers and dynamic acoustic features. Retrieved January 4, 2023, from https://ieeexplore.ieee.org/abstract/document/8128906/?casa_token=NWiFquymi1AAAAAA:LvcZXX_ ZNfYO6CQtEYMaDv4oFti-tTs5PvqszhjpsswTg94Bwhlw5vWAQncPwaiiMJCFZVmXdpQ
  48. Zhang, J., Zeng, X., Wang, M., Pan, Y., Liu, L., Liu, Y., Ding, Y., & Fan, C. (2019). FReeNet: Multi-identity face reenactment. Retrieved from http://arxiv.org/ abs/1905.11805 · arXiv
  49. Zhang, J. X., Ling, Z. H., & Dai, L. R. (2020). Non-parallel sequence-to-sequence voice conversion with disentangled linguistic and speaker representations. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 28, 540-552. https://doi.org/10.1109/TASLP.2019.2960721 · DOI
  50. Zhang, J. X., Ling, Z. H., Liu, L. J., Jiang, Y., & Dai, L. R. (2019). Sequence-to-sequence acoustic modeling for voice conversion. IEEE/ACM Transactions on Audio, Speech, and Language Processing, 27(3), 631-644. https://doi.org/10.1109/TASLP.2019.2892235 · DOI
  51. Zheng, R., & Zhang, S. Text-independent speaker identification using gmm-ubm and frame level likelihood normalization. Retrieved January 4, 2023, from https://ieeexplore.ieee.org/abstract/document/1409643/ AUTHOR BIOGRAPHIES Abhishek Dixit is actively working in the area of audio deepfake detection. He did his Bachelor's in Computer Science and Engineering (Hons.) in Cloud Computing from Chandigarh University in 2021. Presently, he is pursuing Master's in Department of Computer Science and Engi- neering in UIET, Panjab University.
  52. Nirmal Kaur is actively working in the area of energy-efficient and fault-tolerant scheduling on parallel computing systems, and Digital image watermarking techniques. She earned her PhD from IKG-Punjab Technical University, Kapurthala (2017), Masters of Comp. Sc. and Engg. from Punjab Technical University, Kapurthala (2009), and Bachelor of Comp. Sc. and Engg. from Punjabi University, Campus, Patiala (2003), India. Presently, she is an Assistant professor at UIET, Punjab University, Chandigarh, India. She has guided more than 10 M.E. students, and two PhD students are working under her supervision. Staffy Kingra is actively working in the area of digital multimedia forensics with special focus on deepfake detection. She did her Masters in Computer Science and Engineering from UIET, Panjab University, Chandigarh in 2016. Presently, she is a research scholar in Department of Computer Science and Engineering in UIET, Panjab University, Chandigarh, India. Earlier, she has worked as an Assistant Professor in Central University of Punjab, Bathinda. How to cite this article: Dixit, A., Kaur, N., & Kingra, S. (2023). Review of audio deepfake detection techniques: Issues and prospects. Expert Systems, 40(8), e13322. https://doi.org/10.1111/exsy.13322

Cited by

Sorted by citations

Authors
Dixit, A., Kaur, N., & Kingra, S.
Affiliations
Panjab University (IN)
Corresponding author
Kaur, Nirmal
ORCID
Venue
Expert Systems
Publisher
Wiley
Published
2023-04-23
Volume
40
Issue
8
ISSN
  • 0266-4720 (ISSN-L)
  • 1468-0394
DOI
10.1111/exsy.13322
OpenAlex
W4366828058
Cited by
56
References
5
Citations by year
2026: 16 · 2025: 23 · 2024: 14 · 2023: 3
FWCI
5.26
Type
journal article
Language
en
Open access
Licence
Locations
Expert Systems (publishedVersion, public-domain)
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
deepfake audio, deepfake detection, audio generation, generative adversarial networks, variational autoencoder, content manipulation, audio authenticity

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