Transferring Audio Deepfake Detection Capability across Languages
Open accessThe DOI matches its Crossref record; 19 of the 19 references it cites were found in a registry or an index.
Abstract
The proliferation of deepfake content has motivated a surge of detection studies. However, existing detection methods in the audio area exclusively work in English, and there is a lack of data resources in other languages. Cross-lingual deepfake detection, a critical but rarely explored area, urges more study. This paper conducts the first comprehensive study on the cross-lingual perspective of deepfake detection. We observe that English data enriched in deepfake algorithms can teach a detector the knowledge of various spoofing artifacts, contributing to performing detection across language domains. Based on the observation, we first construct a first-of-its-kind cross-lingual evaluation dataset including heterogeneous spoofed speech uttered in the two most widely spoken languages, then explored domain adaptation (DA) techniques to transfer the artifacts detection capability and propose effective and practical DA strategies fitting the cross-lingual scenario. Our adversarial-based DA paradigm teaches the model to learn real/fake knowledge while losing language dependency. Extensive experiments over 137-hour audio clips validate the adapted models can detect fake audio generated by unseen algorithms in the new domain.
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.
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
Davis, S., & Mermelstein, P. (1980) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1980.1163420 · Cited by 5,465
Return of Frustratingly Easy Domain Adaptation
Sun, B., Feng, J., & Saenko, K. (2016) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v30i1.10306 · Cited by 2,004 · cited 1×
“The difference can bemeasured byMMD [26, 28], correlation alignment [34], etc.”
BackgroundASVspoof 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 · cited 2×
“Audio deepfake detection is a part of anti-spoofing research that has been extensively developed with the establishment of ASVspoof challenges [20, 41, 50, 53] where the logical access (LA) track focuses on detecting synthetic speech generated by VC and TTS algorithms [53].”
BackgroundASVspoof 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
Adaptive batch normalization for practical domain adaptation
Li, Y., Wang, N., Shi, J., Hou, X., & Liu, J. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.03.005 · Cited by 601 · cited 1×
“Other approaches modify network structure via batch normalization [22] or ensemble methods [32, 54].”
MethodThe ASVspoof 2017 challenge: assessing the limits of replay spoofing attack detection
Kinnunen, T., Sahidullah, M., Delgado, H., Todisco, M., Evans, N., Yamagishi, J., & Lee, K. A. (2017) · Interspeech 2017 · doi:10.21437/interspeech.2017-1111 · Cited by 535
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 · cited 2×
“Audio deepfake detection is a part of anti-spoofing research that has been extensively developed with the establishment of ASVspoof challenges [20, 41, 50, 53] where the logical access (LA) track focuses on detecting synthetic speech generated by VC and TTS algorithms [53].”
BackgroundEnd-to-End anti-spoofing with RawNet2
Tak, H., Patino, J., Todisco, M., Nautsch, A., Evans, N., & Larcher, A. (2021) · ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp39728.2021.9414234 · Cited by 402 · cited 4×
“[36] modified the original RawNet2[17] to anti-spoofing, which becomes one of the baseline models in ASVspoof2021.”
BackgroundMethodInfluentialDoes Audio Deepfake Detection Generalize?
Müller, N., Czempin, P., Diekmann, F., Froghyar, A., & Böttinger, K. (2022) · Interspeech 2022 · doi:10.21437/interspeech.2022-108 · Cited by 258
One-class learning towards synthetic voice spoofing detection
Zhang, Y., Jiang, F., & Duan, Z. (2021) · IEEE Signal Processing Letters · doi:10.1109/lsp.2021.3076358 · Cited by 241
Countering Malicious DeepFakes: Survey, Battleground, and Horizon
Juefei-Xu, F., Wang, R., Huang, Y., Guo, Q., Ma, L., & Liu, Y. (2022) · International Journal of Computer Vision · doi:10.1007/s11263-022-01606-8 · Cited by 191
Investigating Self-Supervised Front Ends for Speech Spoofing Countermeasures
Wang, X., & Yamagishi, J. (2022) · The Speaker and Language Recognition Workshop (Odyssey 2022) · doi:10.21437/odyssey.2022-14 · Cited by 138
A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection
Wang, X., & Yamagishi, J. (2021) · Interspeech 2021 · doi:10.21437/interspeech.2021-702 · Cited by 137 · cited 1×
“Multiple self-supervised feature extractors were analyzed in [46, 52], while some works focus on building the classifiers[21, 45].”
BackgroundJung, J.-W., Kim, S.-B., Shim, H.-J., Kim, J.-H., & Yu, H.-J. (2020) · Interspeech 2020 · doi:10.21437/interspeech.2020-1011 · Cited by 109
CNN Detection of GAN-Generated Face Images based on Cross-Band Co-occurrences Analysis
Barni, M., Kallas, K., Nowroozi, E., & Tondi, B. (2020) · 2020 IEEE International Workshop on Information Forensics and Security (WIFS) · doi:10.1109/wifs49906.2020.9360905 · Cited by 103
FakeLocator: Robust Localization of GAN-Based Face Manipulations
Huang, Y., Juefei-Xu, F., Guo, Q., Liu, Y., & Pu, G. (2022) · IEEE Transactions on Information Forensics and Security · doi:10.1109/tifs.2022.3141262 · Cited by 102
Das, R. K. (2021) · 2021 Edition of the Automatic Speaker Verification and Spoofing Countermeasures Challenge · doi:10.21437/asvspoof.2021-5 · Cited by 30 · cited 1×
“The research community has widely acknowledged the limitation and proposed approaches to improve model generalization, such as data augmentation [4, 18, 56], domain adversarial learning [44], one-class learning [55], new model architecture [19, 48], and datasets targeting generalization [8, 19].”
BackgroundFMFCC-A: A Challenging Mandarin Dataset for Synthetic Speech Detection
Zhang, Z., Gu, Y., Yi, X., & Zhao, X. (2022) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-95398-0_9 · Cited by 24
A Comparative Study on Recent Neural Spoofing Countermeasures for Synthetic Speech Detection
Wang, X., & Yamagishi, J. (2021) · arXiv · doi:10.48550/arxiv.2103.11326 · Cited by 4
Cited by
Deepfake: definitions, performance metrics and standards, datasets, and a meta-review
Altuncu, E., Franqueira, V. N. L., & Li, S. (2024) · Frontiers in Big Data · doi:10.3389/fdata.2024.1400024 · Cited by 134
“DEepfake CROss-lingual dataset (Ba et al., 2023a,b): This dataset consists of English and Chinese speech samples.”
A lightweight feature extraction technique for deepfake audio detection
Chakravarty, N., & Dua, M. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-18217-9 · Cited by 40
“The experiments described in this work has been based on the ASVspoof 2019 [3] that is used to train the model as it contains the fake audios which are near to human voice and, ASVspoof 2021 [5] dataset has been used to evaluate the trained model against deep-fake audios as it contains deepfake…”
DFIL: Deepfake Incremental Learning by Exploiting Domain-invariant Forgery Clues
Pan, K., Yin, Y., Wei, Y., Lin, F., Ba, Z., Liu, Z., Wang, Z., Cavallaro, L., & Ren, K. (2023) · Proceedings of the 31st ACM International Conference on Multimedia · doi:10.1145/3581783.3612377 · Cited by 37
Self-citation"Better Be Computer or I'm Dumb": A Large-Scale Evaluation of Humans as Audio Deepfake Detectors
Warren, K., Tucker, T., Crowder, A., Olszewski, D., Lu, A., Fedele, C., Pasternak, M., Layton, S., Butler, K., Gates, C., et al. (2024) · Proceedings of the 2024 on ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3658644.3670325 · Cited by 15
Fooling the Forgers: A Multi-Stage Framework for Audio Deepfake Detection
Kashyap, G. S., Siddiqui, Z. H., Azeez, M. A., Ali, R., Kumar, S., Kamuni, N., & Gao, J. (2025) · ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49660.2025.10888175 · Cited by 4
Metadata
- Authors
- Ba, Z., Wen, Q., Cheng, P., Wang, Y., Lin, F., Lu, L., & Liu, Z.
- Venue
- Proceedings of the ACM Web Conference 2023
- Publisher
- ACM
- Published
- 2023-04-30
- Pages
- 2033-2044
- OpenAlex
- W4367046815
- Cited by
- 32
- References
- 19
- FWCI
- 6.05
- Type
- conference paper
- Language
- en
- Open access
- bronze
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.