The DOI matches its Crossref record; 156 of the 156 references it cites were found in a registry or an index; 68 of the 142 sources its claims cite were checked for support.
Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead
Open accessAbstract
Advancements in audio synthesis and manipulation technologies have reshaped applications such as personalised virtual assistants, voice cloning for creative content, and language learning tools. However, the misuse of these technologies to create audio deepfakes has raised serious concerns about security, privacy, and trust. Studies reveal that human judgement of deepfake audio is not always reliable, highlighting the urgent need for robust detection technologies to mitigate these risks. This paper provides a comprehensive survey of recent advancements in audio deepfake detection, with a focus on cutting-edge developments in the past few years. It begins by exploring the foundational methods of audio deepfake generation, including text-to-speech (TTS) and voice conversion (VC), followed by a review of datasets driving progress in the field. The survey then delves into detection approaches, covering frontend feature extraction, backend classification models, and end-to-end systems. Additionally, emerging topics such as privacy-preserving detection, explainability, and fairness are discussed. Finally, this paper identifies key challenges and outlines future directions for developing robust and scalable audio deepfake detection systems.
Overview
CoChat Summary
TL;DR
This survey reviews recent audio deepfake detection, covering generation methods (TTS and VC), datasets, frontend/backend detection approaches, evaluation metrics, and emerging topics such as privacy, fairness, explainability, anti-spoofing, and compression robustness. It concludes that detection has advanced substantially but still faces major gaps in generalisation, lightweight deployment, dataset diversity, and ethical safeguards.
Key findings
- Human judgement of deepfake audio is unreliable: one cited study found humans correctly identified deepfake audio with only 73% accuracy, and a large-scale user study with over 1200 participants found human performance remains suboptimal in complex scenarios.
- Detection research is shifting toward self-supervised learning (SSL) frontends such as Wav2Vec 2.0, WavLM, XLSR (53 languages), and XLS-R (128 languages), with SSL features often outperforming traditional handcrafted features like LFCC and CQCC, especially for unseen attacks.
- Emerging defences report strong results: SafeEar forces ASR models to exceed 90% word error rate in adversarial content-recovery scenarios; AntiFake achieves over 95% protection across five speech synthesisers and three speaker verification systems; Timbre Watermarking reaches 100% extraction accuracy against preprocessing attacks and nearly 90% under extreme low-bitrate/low-pass conditions.
- Dataset coverage is expanding but remains limited: ASVspoof editions include 106 speakers (2015), 42 speakers/177 replay sessions/61 replay configurations (2017), and 107 speakers (2019); newer datasets include MLAAD with 38 languages and 82 spoofing models, and CVoiceFake with five languages.
Methods
- Design: comprehensive survey of recent audio deepfake detection, focusing on cutting-edge developments in the past few years.
- Scope: reviews audio deepfake generation via text-to-speech (TTS) and voice conversion (VC), datasets, evaluation metrics (EER and t-DCF), frontend feature extraction, backend classification models, end-to-end systems, and emerging topics including privacy, fairness, adaptability, explainability, anti-spoofing, and compression robustness.
- Analysis: thematic synthesis and comparative discussion of cited systems, with performance tables such as Table 5/Table 6 on ASVspoof systems ranked by EER and Table 7 on generalisation on the ITW dataset; no primary experimental study is conducted.
Limitations
- The paper states that current detection models generalise poorly across languages, domains, and novel spoofing attacks, and that high-performing systems often depend on large-scale deep learning architectures unsuitable for resource-constrained settings.
- The paper states that existing datasets lack sufficient coverage of real-world scenarios, including diverse languages, accents, background noises, and device conditions, and that imbalanced speaker-type distributions can create fairness concerns.
- Evident from the text: as a survey, it does not provide new empirical validation, and its performance claims depend on the cited works and selected tables; the provided text also does not describe a systematic literature-search protocol.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Figures


Verification
Claims the paper makes
49 Valid · 9 Not valid · 53 Not checked · 31 Own results and interpretations
Every source the paper cites for this claim was checked and supports it.
In the paper
“A study [26] found that humans could correctly identify deepfake audio with an accuracy of only 73%.”
§ Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead > 1. Introduction
What the sources say
- Supportedcites [26] Mai, K.T., Bray, S., Davies, T., and Griffin, L.D. (2023). Warning: Humans cannot reliably detect speech deepfakes. PLoS ONE, 18. sourceExperimental — accuracy not yet measured
The Results (Overall performance) report that participants identified deepfake clips with 73% accuracy in the unary task. The figure applies to that task; accuracy was higher when participants compared paired clips. (LLM, conf=0.94)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Another large-scale user study [27] involving over 1200 participants showed that while humans can rely on intuition and linguistic features for classification, their overall performance remains suboptimal in certain scenarios, particularly in complex environments.”
§ Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead > 1. Introduction
What the sources say
- Partly supportedcites [27] Warren, K., Tucker, T., Crowder, A., Olszewski, D., Lu, A., Fedele, C., Pasternak, M., Layton, S., Butler, K., and Gates, C. (2024, January 14–18). “Better Be Computer or I’m Dumb”: A Large-Scale Evaluation of Humans as Audio Deepfake Detectors. Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security (CCS), Salt Lake City, UT, USA. sourceExperimental — accuracy not yet measured
The Introduction reports a study of over 1,200 participants with 73% classification accuracy, and §4.1 reports lower accuracy on some datasets. The paper also says participants rely on linguistic features and intuition (Abstract; Finding 3). It does not establish that performance is particularly suboptimal in complex environments; the tested conditions were samples from three audio-deepfake datasets, and no page numbers are available in our copy. (LLM, conf=0.84)
plainn = 1,200- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Unlike deepfake detection for images and videos, which has been intensively studied [28,29,30,31], audio deepfake detection has been overlooked in most of the existing surveys until recently [32].”
§ Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead > 1. Introduction
What the sources say
- Supportedcites [32] Khanjani, Z., Watson, G., and Janeja, V.P. (2023). Audio deepfakes: A survey. Front. Big Data, 5. source
The Abstract says most existing surveys focus on image and video deepfakes and overlook audio deepfakes. The Introduction likewise identifies the lack of attention to audio deepfakes in surveys as the motivation for this survey, which covers their detection. (LLM, conf=0.93)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Furthermore, while existing machine learning models have achieved considerable progress, they have not consistently outperformed human intuition [27].”
§ Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead > 1. Introduction
What the sources say
- Supportedcites [27] Warren, K., Tucker, T., Crowder, A., Olszewski, D., Lu, A., Fedele, C., Pasternak, M., Layton, S., Butler, K., and Gates, C. (2024, January 14–18). “Better Be Computer or I’m Dumb”: A Large-Scale Evaluation of Humans as Audio Deepfake Detectors. Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security (CCS), Salt Lake City, UT, USA. source
Finding 6 states that models do not strictly perform better than humans in classifying audio samples. Finding 3 says participants rely heavily on intuition when distinguishing deepfake audio. Together, these support the paraphrase, though the paper compares overall human performance rather than intuition alone. (LLM, conf=0.94)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“These constraints hinder the models’ ability to generalise to real-world deepfake audio.”
§ Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead > 3. Dataset > 3.3. Additional Datasets
What the sources say
- Partly supportedcites [64] Müller, N.M., Czempin, P., Dieckmann, F., Froghyar, A., and Böttinger, K. (2022). Does audio deepfake detection generalize?. arXiv. sourceExperimental — accuracy not yet measured
The Introduction identifies ASVspoof’s limited synthesis algorithms and studio-recorded source audio. Section 5.3 reports substantially worse model performance on in-the-wild audio, but the Conclusion says only that models may have been tailored too closely to ASVspoof. The paper does not establish that those specific dataset constraints caused the loss of generalisation. (LLM, conf=0.84)
plainstates a cause- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
52 more claims · 23 with their cited sources checked
Create a free account to see all 142 claims and to verify the 9 references still open.
- Probably not a substantial claim. 30
- It may state two separate claims as one. 66
Citations
Cites
Very Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707
A Comprehensive Survey on Graph Neural Networks
Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., & Yu, P. S. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2020.2978386 · Cited by 11,000
WaveNet: a generative model for raw audio
Oord, A. V. D., Dieleman, S., Zen, H., Simonyan, K., Vinyals, O., Graves, A., Kalchbrenner, N., Senior, A., & Kavukcuoglu, K. (2016) · arXiv · doi:10.48550/arxiv.1609.03499 · Cited by 4,911
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
WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing
Chen, S., Wang, C., Chen, Z., Wu, Y., Liu, S., Chen, Z., Li, J., Kanda, N., Yoshioka, T., Xiao, X., et al. (2022) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2022.3188113 · Cited by 1,926
Hinton, G. E., Krizhevsky, A., & Wang, S. D. (2011) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-21735-7_6 · Cited by 1,360
AST: Audio Spectrogram Transformer
Gong, Y., Chung, Y.-A., & Glass, J. (2021) · Interspeech 2021 · doi:10.21437/interspeech.2021-698 · Cited by 1,159
The Emergence of Deepfake Technology: A Review
Westerlund, M. (2019) · Technology Innovation Management Review · doi:10.22215/timreview/1282 · Cited by 1,138
Multi-attentional Deepfake Detection
Zhao, H., Wei, T., Zhou, W., Zhang, W., Chen, D., & Yu, N. (2021) · 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr46437.2021.00222 · Cited by 1,019
Research works printed but not linked: 15
- Ubert, J. (2023). Fake it: Attacking Privacy Through Exploiting Digital Assistants Using Voice Deepfakes. [Ph.D. Thesis, Marymount University].
- Ren, 2021, A survey on speech forgery and detection, J. Signal Process., 37, 2412
- Ren, Y., Hu, C., Tan, X., Qin, T., Zhao, S., Zhao, Z., and Liu, T.Y. (2020). Fastspeech 2: Fast and high-quality end-to-end text to speech. arXiv.
- Kong, Z., Ping, W., Huang, J., Zhao, K., and Catanzaro, B. (2020). Diffwave: A versatile diffusion model for audio synthesis. arXiv.
- Lee, S.G., Ping, W., Ginsburg, B., Catanzaro, B., and Yoon, S. (2022). Bigvgan: A universal neural vocoder with large-scale training. arXiv.
- Lee, S.H., Choi, H.Y., and Lee, S.W. (2024). Accelerating High-Fidelity Waveform Generation via Adversarial Flow Matching Optimization. arXiv.
- Wang, D., and Zhang, X. (2015). Thchs-30: A free chinese speech corpus. arXiv.
- Frank, J., and Schönherr, L. (2021). Wavefake: A data set to facilitate audio deepfake detection. arXiv.
- Baevski, 2020, wav2vec 2.0: A framework for self-supervised learning of speech representations, Adv. Neural Inf. Process. Syst., 33, 12449
- Iqbal, F., Abbasi, A., Javed, A.R., Jalil, Z., and Al-Karaki, J.N. (2022, January 17–21). Deepfake Audio Detection Via Feature Engineering And Machine Learning. Proceedings of the CIKM 2022 Workshops, Atlanta, GA, USA. CEUR Workshop Proceedings; 2022.
- Alexey, D. (2020). An image is worth 16 × 16 words: Transformers for image recognition at scale. arXiv.
- Kipf, T.N., and Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv.
- Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (2017). Graph attention networks. arXiv.
- Zhang, X., Yi, J., and Tao, J. (2024). EVDA: Evolving Deepfake Audio Detection Continual Learning Benchmark. arXiv.
- Channing, G., Sock, J., Clark, R., Torr, P., and de Witt, C.S. (2024). Toward Robust Real-World Audio Deepfake Detection: Closing the Explainability Gap. arXiv.
Cited by
Deepfake Audio Detection in Voice Authentication: A Spectral and CNN-Based Comprehensive Review
Salih, A. O. M., Emam, A. H. M., Ahmed, A. B. G. E., Khalifa, M., Suliman, A., & Babiker, N. B. M. (2025) · Engineering Technology & Applied Science Research · doi:10.48084/etasr.13400 · Cited by 17
Kheir, Y. E., Das, A., Erdogan, E. E., Ritter-Guttierez, F., Polzehl, T., & Möller, S. (2025) · IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA) · doi:10.1109/waspaa66052.2025.11230938 · Cited by 12
A Multimodal Voice Phishing Detection System Integrating Text and Audio Analysis
Kim, J., Gu, S., Kim, Y.-M., Lee, S., & Kang, C. (2025) · Applied Sciences · doi:10.3390/app152011170 · Cited by 5
MOSS: Multi-modal Source Separation and Feature Fusion for Music Deepfake Detection
Zhu, X., Li, W., Qiao, M., Yang, Z., & Wang, Y. (2026) · Lecture notes in computer science · doi:10.1007/978-981-95-6960-1_38 · Cited by 2
Metadata
- Authors
- Zhang, B., Cui, H., Nguyen, V., & Whitty, M.
- Affiliations
- Monash University (AU)
- Corresponding author
- Cui, Hui
- Venue
- Sensors
- Publisher
- MDPI AG
- Published
- 2025-03-22
- Volume
- 25
- Issue
- 7
- Pages
- 1989
- ISSN
- 1424-8220 (ISSN-L)
- PMC
- PMC11991371
- OpenAlex
- W4408799803
- Cited by
- 72
- References
- 145
- Citations by year
- 2026: 54 · 2025: 18
- FWCI
- 6.87
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- Locations
- Sensors (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- Monash University Research Portal (Monash University) (submittedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Keywords
- audio deepfake detection, deepfake audio, voice cloning, text-to-speech, voice conversion, sound synthesis, deepfake datasets, privacy-preserving detection, explainability, algorithmic fairness, sound manipulation
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-09-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-09-30 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead. https://app.cochat.ai/papers/10.3390/s25071989
- The paper
- Zhang, B., Cui, H., Nguyen, V., & Whitty, M. (2025). Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead. Sensors. https://doi.org/10.3390/s25071989
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.