Generalization of Audio Deepfake Detection
The DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Audio Deepfakes, technically known as logical-access voice spoofing techniques, have become an increased threat on voice interfaces due to the recent breakthroughs in speech synthesis and voice conversion technologies. Effectively detecting these attacks is critical to many speech applications including automatic speaker verification systems. As new types of speech synthesis and voice conversion techniques are emerging rapidly, the generalization ability of spoofing countermeasures is becoming an increasingly critical challenge. This paper focuses on overcoming this issue by using large margin cosine loss function (LMCL) and online frequency masking augmentation to force the neural network to learn more robust feature embeddings. We evaluate the performance of the proposed system on the ASVspoof 2019 logical access (LA) dataset. Additionally, we evaluate it on a noisy version of the ASVspoof 2019 dataset using publicly available noises to simulate more realistic scenarios. Finally, we evaluate the proposed system on a copy of the dataset that is logically replayed through the telephony channel to simulate spoofing attacks in the call center scenario. Our baseline system is based on residual neural network, and has achieved the lowest equal error rate (EER) of 4.04% among all single-system submissions during the ASVspoof 2019 challenge. Furthermore, the additional improvements proposed in this paper reduce the EER to 1.26%.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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?
In the citing paper
“Chen et al. [102] employed augmented LFB features to train a deep residual network (ResNet), resulting in a model with strong generalisation capabilities”
From Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead (2025) · cites this paper as [102]
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“The Residual Network is a variant of the ResNet18 described in [25] where the global average pooling (GAP) layer is replaced by mean and standard deviation pooling layers [26].”
MethodWaveNet: 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 · cited 1×
“WaveNet [2], proposed in 2016, was the first end-to-end speech synthesizer that directly uses the raw audio for training, and showed a mean opinion score (MOS) very close to human speech.”
BackgroundSpecAugment: A Simple Data Augmentation Method for Automatic Speech Recognition
Park, D. S., Chan, W., Zhang, Y., Chiu, C.-C., Zoph, B., Cubuk, E. D., & Le, Q. V. (2019) · Interspeech 2019 · doi:10.21437/interspeech.2019-2680 · Cited by 3,558 · cited 1×
“Additionally, inspired by SpecAugment [17], we propose to add FreqAugment , a layer that randomly masks adjacent frequency channels during the DNN training, to further increase the generalization ability of the DNN model.”
MethodX-Vectors: Robust DNN Embeddings for Speaker Recognition
Snyder, D., Garcia-Romero, D., Sell, G., Povey, D., & Khudanpur, S. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8461375 · Cited by 2,620 · cited 1×
“The Residual Network is a variant of the ResNet18 described in [25] where the global average pooling (GAP) layer is replaced by mean and standard deviation pooling layers [26].”
MethodA 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×
“pulse responses (RIR) for reverberation were chosen from publicly available RIR datasets(2) [18, 19, 20].”
MethodPerceptual 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
A binaural room impulse response database for the evaluation of dereverberation algorithms
Jeub, M., Schafer, M., & Vary, P. (2009) · 2009 16th International Conference on Digital Signal Processing · doi:10.1109/icdsp.2009.5201259 · Cited by 391 · cited 1×
“In order to evaluate our system under noisy conditions, data augmentation is performed on original ASVspoof 2019 dataset by modifying the the data augmentation technique from Kaldi. pulse responses (RIR) for reverberation were chosen from publicly available RIR datasets 2 [18, 19, 20].”
MethodA comparison of features for synthetic speech detection
Sahidullah, M., Kinnunen, T., & Hanilçi, C. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-472 · Cited by 382 · cited 1×
“Similar filter bank based cepstral features such as linear frequency cepstral coef-ficients (LFCCs) also showed competitive performance in synthetic speech detection [22] and speaker recognition [23].”
BackgroundDeep Voice: Real-time Neural Text-to-Speech
Arık, S. Ö., Chrzanowski, M., Coates, A., Diamos, G., Gibiansky, A., Kang, Y., Li, X., Miller, J. J., Ng, A. Y., Raiman, J., et al. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1702.07825 · Cited by 381 · cited 1×
“Similar quality was shown by other TTS systems such as Deep Voice [3] and Tacotron [4], and also by VC systems [5, 6].”
Kinnunen, T., Lorenzo-Trueba, J., Yamagishi, J., Toda, T., Saito, D., Villavicencio, F., & Ling, Z. (2018) · The Speaker and Language Recognition Workshop (Odyssey 2018) · doi:10.21437/odyssey.2018-27 · Cited by 23
Cited by
Masood, M., Nawaz, M., Malik, K. M., Javed, A., Irtaza, A., & Malik, H. (2023) · Applied Intelligence · doi:10.1007/s10489-022-03766-z · Cited by 543
“[213] introduced a technique to identify audio manipulation.”
MethodInfluentialAASIST: Audio Anti-Spoofing Using Integrated Spectro-Temporal Graph Attention Networks
Jung, J.-W., Heo, H.-S., Tak, H., Shim, H.-J., Chung, J. S., Lee, B.-J., Yu, H.-J., & Evans, N. (2022) · ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp43922.2022.9747766 · Cited by 399
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
“[15] proposed a system which employs the ResNet with large margin cosine loss and applied frequency mask augmentation.”
MethodTak, H., Jung, J.-W., Patino, J., Kamble, M., Todisco, M., & Evans, N. (2021) · 2021 Edition of the Automatic Speaker Verification and Spoofing Countermeasures Challenge · doi:10.21437/asvspoof.2021-1 · Cited by 197
A Review of Modern Audio Deepfake Detection Methods: Challenges and Future Directions
Almutairi, Z., & Elgibreen, H. (2022) · Algorithms · doi:10.3390/a15050155 · Cited by 187
“People can use it as a logical-access voice spoofing technique [4], where it can be used to manipulate public opinion for propaganda, defamation, or even terrorism.”
Metadata
- Authors
- Chen, T., Kumar, A., Nagarsheth, P., Sivaraman, G., & Khoury, E.
- Venue
- The Speaker and Language Recognition Workshop (Odyssey 2020)
- Publisher
- ISCA
- Published
- 2020-11-01
- Pages
- 132-137
- OpenAlex
- W3024920698
- Cited by
- 200
- References
- 10
- Citations by year
- 2026: 16 · 2025: 35 · 2024: 47 · 2023: 38 · 2022: 35 · 2021: 26 · 2020: 3
- FWCI
- 7.87
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- The Speaker and Language Recognition Workshop (Odyssey) (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- audio deepfake detection, audio forensics, synthetic speech detection
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Generalization of Audio Deepfake Detection. https://app.cochat.ai/papers/10.21437/odyssey.2020-19
- The paper
- Chen, T., Kumar, A., Nagarsheth, P., Sivaraman, G., & Khoury, E. (2020). Generalization of Audio Deepfake Detection. The Speaker and Language Recognition Workshop (Odyssey 2020). https://doi.org/10.21437/odyssey.2020-19
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.