Mel-Spectrogram Image-Based End-to-End Audio Deepfake Detection Under Channel-Mismatched Conditions
The DOI matches its Crossref record; 34 of the 37 references it cites were found in a registry or an index.
Abstract
This work focuses on the problem of detecting fake audio clips. To improve current audio spoofing detection models, we propose a selection of multiple audio augmentations spe-cially designed to resemble audio spoofing attacks. These augmentations are experimentally found to be very useful and using them achieves a notable performance of 2.8% EER on the ASVspoof 2019 challenge evaluation set. Unlike the widely employed acoustic features, in this paper we explore the use of Mel-spectrogram image features and employ vari-ous audio codecs to achieve robustness to codec and transmission channel variability present in the ASVspoof2021 Evalu-ation set. To better handle spectral information, crucial to de-tect spoofing, we adopt the WaveletCNN and VGG16 archi-tectures which outperform all baselines. Finally, we find that robustness of countermeasure systems degrades dramatically when provided with speech samples degraded through VoIP network transmission or mismatching audio compression.
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
“Fathan et al. [104] explored the use of Mel Spectrograms combined with VGG16 and WaveletCNN architectures, significantly outperforming traditional cepstral features in detecting LA attacks.”
From Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead (2025) · cites this paper as [104]
In the citing paper
“Fathan et al. [104] used VGG16 as a feature extractor to analyse spectrograms, focusing on capturing artefacts in the frequency and temporal dimensions. Their findings demonstrated the potential of VGG16 combined with Mel-Spectrograms for detecting fake audio, particularly under complex scenarios such as channel mismatch”
From Audio Deepfake Detection: What Has Been Achieved and What Lies Ahead (2025) · cites this paper as [104]
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 · cited 3×
“Since the VGG16 architecture [4] is proposed for image recognition, it uses 2D-CNN layers to capture the spatial correlation within the input image.”
MethodInfluentialWaveNet: 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×
“The ASVspoof2019 dataset [3] includes bonafide (legitimate), synthetic speech, replay attack, and converted voice signals generated with recent algorithms such as Tacotron2 [22] and Wavenet [23].”
MethodNatural 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
CNN architectures for large-scale audio classification
Hershey, S., Chaudhuri, S., Ellis, D. P. W., Gemmeke, J. F., Jansen, A., Moore, R. C., Plakal, M., Platt, D., Saurous, R. A., Seybold, B., et al. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7952132 · Cited by 2,023
Additive Margin Softmax for Face Verification
Wang, F., Cheng, J., Liu, W., & Liu, H. (2018) · IEEE Signal Processing Letters · doi:10.1109/lsp.2018.2822810 · Cited by 1,106 · cited 1×
“In order to improve performance on previously unseen data and to generalize to out-of-domain spoofed speech samples, we use the additive margin softmax loss (AM Softmax) [30] for training.”
MethodAutoAugment: Learning Augmentation Policies from Data
Cubuk, E. D., Zoph, B., Mane, D., Vasudevan, V., & Le, Q. V. (2018) · arXiv · doi:10.48550/arxiv.1805.09501 · Cited by 1,106 · cited 1×
“Additionally, there have recently been some works done in computer vision to automatize the selection of augmentations for supervised [10] and unsupervised models [11].”
BackgroundAudio augmentation for speech recognition
Ko, T., Peddinti, V., Povey, D., & Khudanpur, S. (2015) · Interspeech 2015 · doi:10.21437/interspeech.2015-711 · Cited by 1,104 · cited 1×
“Therefore, data augmentation is an important part of audio tasks and numerous papers have focused on it [8,9].”
BackgroundState-of-the-Art Speech Recognition with Sequence-to-Sequence Models
Chiu, C.-C., Sainath, T. N., Wu, Y., Prabhavalkar, R., Nguyen, P., Chen, Z., Kannan, A., Weiss, R. J., Rao, K., Gonina, E., et al. (2018) · 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2018.8462105 · Cited by 1,043
A Theory for Multiresolution Signal Decomposition: The Wavelet Representation
Mallat, S. G. (2009) · Fundamental Papers in Wavelet Theory · doi:10.1515/9781400827268.494 · Cited by 1,018
Converting video formats with FFmpeg
Tomar, S. (2006) · Linux journal · doi:10.5555/1134782.1134792 · Cited by 647
Research works printed but not linked: 5
- Fujieda, 2018, Wavelet convolutional neural net-works, arXiv preprint
- 2019, ASVspoof consortium
- 10.1109/ICASSP.2018.8461852 · DOI
- Kang, 2021, Crim’s system description for the asvspoof 2021 challenge
- 2021, opencore-amr dev
Cited by
ASVspoof 2021: Towards Spoofed and Deepfake Speech Detection in the Wild
Liu, X., Wang, X., Sahidullah, M., Patino, J., Delgado, H., Kinnunen, T., Todisco, M., Yamagishi, J., Evans, N., Nautsch, A., et al. (2023) · IEEE/ACM Transactions on Audio, Speech, and Language Processing · doi:10.1109/taslp.2023.3285283 · Cited by 312
“Investigation on DA techniques applied on anti-spoofing can be found in [44], [45], [47].”
BackgroundAudio 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
“Fathan et al. [104] explored the use of Mel Spectrograms combined with VGG16 and WaveletCNN architectures, significantly out-performing traditional cepstral features in detecting LA attacks.”
Deepfakes as a threat to a speaker and facial recognition: An overview of tools and attack vectors
Firc, A., Malinka, K., & Hanáček, P. (2023) · Heliyon · doi:10.1016/j.heliyon.2023.e15090 · Cited by 80
Exploring autonomous methods for deepfake detection: A detailed survey on techniques and evaluation
Sunil, R., Mer, P., Diwan, A., Mahadeva, R., & Sharma, A. (2025) · Heliyon · doi:10.1016/j.heliyon.2025.e42273 · Cited by 54
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
“For this aim, machine learning methods such as Support Vector Machines (SVM) [17], Naive Bayes (NB) [18], K-Nearest Neighbours (KNN) [19], and Random Forest (RF) [20] have been used.”
Metadata
- Authors
- Fathan, A., Alam, J., & Kang, W. H.
- Affiliations
- Computer Research Institute of Montréal (CA)
- Venue
- 2022 IEEE International Conference on Multimedia and Expo (ICME)
- Publisher
- IEEE
- Published
- 2022-07-18
- Pages
- 1-6
- OpenAlex
- W4299806750
- Cited by
- 29
- References
- 37
- Citations by year
- 2025: 13 · 2024: 10 · 2023: 6
- FWCI
- 4.01
- Type
- conference paper
- Language
- en
- Open access
- closed
- Locations
- IEEE International Conference on Multimedia and Expo (ICME) (publishedVersion)
- Funders
- NSERC (RGPIN-2019-05381)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Physical Sciences › Computer Science › Signal Processing › Speech and Audio Processing
- Physical Sciences › Computer Science › Signal Processing › Music and Audio Processing
- Keywords
- audio deepfake detection, audio data augmentation, spoofing attacks, channel mismatch, audio codec, VGG16, ASVspoof 2019, ASVspoof 2021
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: Mel-Spectrogram Image-Based End-to-End Audio Deepfake Detection Under Channel-Mismatched Conditions. https://app.cochat.ai/papers/10.1109/icme52920.2022.9859621
- The paper
- Fathan, A., Alam, J., & Kang, W. H. (2022). Mel-Spectrogram Image-Based End-to-End Audio Deepfake Detection Under Channel-Mismatched Conditions. 2022 IEEE International Conference on Multimedia and Expo (ICME). https://doi.org/10.1109/icme52920.2022.9859621
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.