The DOI matches its Crossref record; 72 of the 72 references it cites were found in a registry or an index; 2 of the 13 sources its claims cite were checked for support.
MLAAD: The Multi-Language Audio Anti-Spoofing Dataset
Open accessAbstract
Text-to-Speech (TTS) technology brings significant advantages, such as giving a voice to those with speech impairments, but also enables audio deepfakes and spoofs. The former mislead individuals and may propagate misinformation, while the latter undermine voice biometric security systems. AI-based detection can help to address these challenges by automatically differentiating between genuine and fabricated voice recordings. However, these models are only as good as their training data, which currently is severely limited due to an overwhelming concentration on English and Chinese audio in anti-spoofing databases, thus restricting its worldwide effectiveness.In response, this paper presents the Multi-Language Audio Anti-Spoof Dataset (MLAAD), created using 52 TTS models, comprising 22 different architectures, to generate 160.2 hours of synthetic voice in 23 different languages. We train and evaluate three state-of-the-art deepfake detection models with MLAAD, and observe that MLAAD demonstrates superior performance over comparable datasets like InTheWild or FakeOrReal when used as a training resource. Furthermore, in comparison with the renowned ASVspoof 2019 dataset, MLAAD proves to be a complementary resource. In tests across eight datasets, MLAAD and ASVspoof 2019 alternately outperformed each other, both excelling on four datasets.By publishing1MLAAD and making trained models accessible via an interactive webserver2, we aim to democratize antispoofing technology, making it accessible beyond the realm of specialists, thus contributing to global efforts against audio spoofing and deepfakes.
Overview
CoChat Summary
TL;DR
MLAAD is a multilingual synthetic-audio anti-spoofing dataset, version 11, containing 1153.6 hours of synthetic voice from 205 TTS models across 54 languages, built as an extension of M-AILABS. Training three state-of-the-art detectors on MLAAD shows cross-dataset generalization that is complementary to ASVspoof 2019, with MLAAD and ASVspoof 2019 each achieving the highest accuracy on four of eight test datasets.
Key findings
- MLAAD v11 comprises 1153.6 hours of synthesized speech, 205 state-of-the-art TTS models, 127 different architectures, and 54 languages; only synthesized samples are distributed, with originals referred to the M-AILABS dataset.
- In cross-dataset evaluation, MLAAD and ASVspoof 2019 alternately outperformed each other, each excelling on four of eight datasets; FakeOrReal and InTheWild did not top any case.
- Models trained on MLAAD were the only ones that did not exhibit performance significantly below 50%, hinting at possibly less detrimental shortcut features in MLAAD than in other datasets.
- Example results: SSL W2V2 trained on MLAAD achieved 78.0±15.3 on ASVspoof19, 73.7±10.4 on ASVspoof21-LA, 66.0±34.2 on Voc.v, and 69.8±38.4 on WaveFake; WhisperDF trained on MLAAD achieved 86.7±29.4 on Voc.v and 97.2±3.3 on WaveFake.
Methods
- Dataset construction: MLAAD extends M-AILABS, which has eight original languages (English, French, German, Italian, Polish, Russian, Spanish, Ukrainian). For each language/TTS-model combination, a random subset of 1000 instances is selected; if the target language is absent from M-AILABS, English texts are translated via Neural Machine Translation. Synthetic versions are generated with TTS models from Coqui.ai and Hugging Face, plus Griffin Lim resynthesis for original languages. Audio is stored at 22,050 Hz, 16-bit depth, WAV format, with meta.csv files.
- Evaluation: three state-of-the-art voice anti-spoofing models—RawGat-ST, SSL-W2V2, and WhisperDF—are trained separately on MLAAD v1, ASVspoof19, FakeOrReal, and In-The-Wild, then evaluated on eight datasets including ASVspoof21-DF, ASVspoof21-LA, Voc.v, and WaveFake. Each configuration is run five times, reporting mean accuracy and standard deviation. Training uses noise/music augmentation and codec encoding, with balanced undersampling per epoch, 100 epochs, and early stopping based on training accuracy.
- Accuracy is used instead of equal error rate (EER) because some test datasets, such as Voc.v, consist only of spoof samples, and accuracy is argued to better suit real-world actionable output.
Limitations
- Stated: evaluation of multilingual anti-spoofing capabilities remains future work because there are no other comparable multilingual datasets against which to evaluate MLAAD; strong results on WaveFake, which includes English and Japanese, suggest models profit from MLAAD’s language diversity.
- Stated: the reported results use MLAAD v1; the evaluation was not re-run for subsequent releases, though later versions are expected to perform substantially better given much larger volume and diversity.
- Stated/evident: models sometimes learn shortcut features that transfer well to some datasets but fail completely on others, with some accuracy scores below 50%; MLAAD-trained models are less affected but the issue is not eliminated.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 5 Not valid · 7 Not checked · 30 Own results and interpretations
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“the increasing prevalence of suspected deepfakes in various non-English media [7], [8]”
What the sources say
- Could not verifycites [7] sourceExperimental — accuracy not yet measured
The Introduction (p. 2) describes a database of spoofed speech and its potential use in fake-audio detection, not suspected deepfakes appearing in media. Searches for non-English media and increasing prevalence found no passage in our copy addressing either point. (LLM, conf=0.73)
- Not matchedcites [8] Content authenticity initiative
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- 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.
- Could not verify
The cited sources that could be checked couldn’t confirm or dispute this claim.
In the paper
“features audio clips in eight original languages: English, French, German, Italian, Polish, Russian, Spanish, and Ukrainian.”
What the sources say
- Could not verifycites [9] sourceExperimental — accuracy not yet measured
The Database section (p. 1) describes ASVspoof 2021 data derived from VCTK and other undisclosed corpora, but does not discuss M-AILABS. Searches for M-AILABS and the listed languages found no passage identifying the corpus or confirming its eight languages. (LLM, conf=0.77)
plain- 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.
- Rests on an assumption not shown here.
- Could not verify
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“no single training dataset consistently outperforms the others”
plainn = 8- 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.
- Rests on 3 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“the resulting models achieve high performance scores in cross-dataset evaluations”
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.
- Rests on 3 assumptions not shown here.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“comprising 127 different architectures”
- Also stated in: § V. CONCLUSION · plain
“composed of 127 architectures”
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.
- Also stated in: § V. CONCLUSION · plain
21 more claims
Create a free account to see all 42 claims and to verify the 2 references still open.
- Probably not a substantial claim. 11
- It may state two separate claims as one. 7
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ REFERENCES
“Each row represents a model trained on one dataset and evaluated on the remaining seven datasets, as described in Section IV-B.”
§ B. Cross-Dataset Evaluation
“Finally, we refrain from using ASVspoof21-LA to assess the crossdataset performance of models trained on ASVspoof19, given the substantial similarity between the two datasets.”
Citations
Cites
Binary Codes capable of currecting deletions, insertions, and reversals
Levenshtein, V. I. (1965) · Soviet physics. Doklady · Cited by 14,807
Transformers: State-of-the-Art Natural Language Processing
Wolf, T., Debut, L., Sanh, V., Chaumond, J., Delangue, C., Moi, A., Cistac, P., Rault, T., Louf, R., Funtowicz, M., et al. (2020) · Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations · doi:10.18653/v1/2020.emnlp-demos.6 · Cited by 8,741
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
Signal estimation from modified short-time Fourier transform
Griffin, D., & Lim, J. (1984) · IEEE Transactions on Acoustics, Speech, and Signal Processing · doi:10.1109/tassp.1984.1164317 · Cited by 2,095
MesoNet: a Compact Facial Video Forgery Detection Network
Afchar, D., Nozick, V., Yamagishi, J., & Echizen, I. (2018) · 2018 IEEE International Workshop on Information Forensics and Security (WIFS) · doi:10.1109/wifs.2018.8630761 · Cited by 1,872
(2018) · doi:10.21437/interspeech.2018-1456 · Cited by 1,366 · cited 1×
“The field is supported by numerous open-source toolkits [37]–[40], which offer access to various speech synthesis methods.”
Common Voice: A Massively-Multilingual Speech Corpus
Ardila, R., Branson, M., Davis, K., Henretty, M., Kohler, M., Meyer, J., Morais, R., Saunders, L., Tyers, F. M., & Weber, G. (2019) · arXiv · doi:10.48550/arxiv.1912.06670 · Cited by 905 · cited 2×
“The overall composition of TTS training datasets is highly diverse and includes both professional speakers [41] as well as audio recorded in community-driven initiatives [44].”
A study on data augmentation of reverberant speech for robust speech recognition
Ko, T., Peddinti, V., Povey, D., Seltzer, M. L., & Khudanpur, S. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7953152 · Cited by 886 · cited 1×
“We augment all training data with random noise sourced from the RIRS Noises [66] and Noise ESC-50 [67] datasets, along with music randomly selected from the Free Music Archive (Instrumental Music) [68] and Musan [69], overlaying randomly chosen noise or music with a 5% probability per sample.”
Perceptual Signal Analysis and Eigenvalue-Weighted Metrics for Musical Audio Quality Assessment
Majumder, S. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-27827-2_35 · Cited by 845
ASVspoof 2019: A large-scale public database of synthesized, converted and replayed speech
Wang, X., Yamagishi, J., Todisco, M., Delgado, H., Nautsch, A., Evans, N., Sahidullah, M., Vestman, V., Kinnunen, T., Lee, K. A., et al. (2020) · Computer Speech & Language · doi:10.1016/j.csl.2020.101114 · Cited by 505 · cited 4×
“Contrary to typical voice anti-spoofing evaluations [7], [9], [59], [65] that use the equal error rate (EER), we opt for accuracy as our evaluation metric.”
Influential
Research works printed but not linked: 1
- 10.21437/Odyssey.2022-16 · DOI · Paper page
Cited by
A Survey on Speech Deepfake Detection
Li, M., Ahmadiadli, Y., & Zhang, X.-P. (2025) · ACM Computing Surveys · doi:10.1145/3714458 · Cited by 101
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
Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead
Zhang, B., Cui, H., Nguyen, V., & Whitty, M. (2025) · Sensors · doi:10.3390/s25071989 · Cited by 72
“Similarly, Müller et al. [91] developed the Multi-Language Audio Anti-Spoofing Dataset (MLAAD), a large-scale multilingual dataset covering 38 languages and 82 spoofing models, offering robust support for the global applicability of audio deepfake detection systems.”
The Codecfake Dataset and Countermeasures for the Universally Detection of Deepfake Audio
Xie, Y., Lu, Y., Fu, R., Wen, Z., Wang, Z., Tao, J., Qi, X., Wang, X., Liu, Y., Cheng, H., et al. (2025) · IEEE Transactions on Audio Speech and Language Processing · doi:10.1109/taslpro.2025.3525966 · Cited by 111
“Consequently, our constructed dataset incorporates these three meth-ods among the seven approaches.”
SpoofCeleb: Speech Deepfake Detection and SASV in the Wild
Jung, J.-W., Wu, Y., Wang, X., Kim, J.-H., Maiti, S., Matsunaga, Y., Shim, H.-J., Tian, J., Evans, N., Chung, J. S., et al. (2025) · IEEE Open Journal of Signal Processing · doi:10.1109/ojsp.2025.3529377 · Cited by 80
Metadata
- Authors
- Müller, N. M., Kawa, P., Choong, W. H., Casanova, E., Gölge, E., Müller, T., Syga, P., Sperl, P., & Böttinger, K.
- Affiliations
- Fraunhofer Institute for Applied and Integrated Security (DE)
- Wrocław University of Science and Technology (PL)
- Coqui.ai
- Thorsten-Voice
- Venue
- 2024 International Joint Conference on Neural Networks (IJCNN)
- Publisher
- IEEE
- Published
- 2024-06-30
- Pages
- 1-7
- ISSN
- 2161-4407 (ISSN-L)
- OpenAlex
- W4402351686
- Cited by
- 73
- References
- 66
- Citations by year
- 2026: 21 · 2025: 41 · 2024: 11
- FWCI
- 21.58
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Locations
- International Joint Conference on Neural Networks (IJCNN) (publishedVersion)
- arXiv (Cornell University) (submittedVersion)
- Fraunhofer-Publica (Fraunhofer-Gesellschaft) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Ministry of Economic Affairs
- 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 › Speech and Audio Processing
- Keywords
- audio deepfake detection, text-to-speech, multilingual speech, misinformation, ASVspoof 2019
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 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-10 (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: MLAAD: The Multi-Language Audio Anti-Spoofing Dataset. https://app.cochat.ai/papers/10.1109/ijcnn60899.2024.10650962
- The paper
- Müller, N. M., Kawa, P., Choong, W. H., Casanova, E., Gölge, E., Müller, T., Syga, P., Sperl, P., & Böttinger, K. (2024). MLAAD: The Multi-Language Audio Anti-Spoofing Dataset. 2024 International Joint Conference on Neural Networks (IJCNN). https://doi.org/10.1109/ijcnn60899.2024.10650962
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.