CoChat
Sign Up
Cited by 30Speech Recognition and SynthesisMusic and Audio ProcessingDigital Media Forensic Detection

An Initial Investigation for Detecting Vocoder Fingerprints of Fake Audio

Open access
Yan, Xinrui, Yi, Jiangyan, Tao, Jianhua, Wang, Chenglong, Ma, Haoxin, Wang, Tao, Wang, Shiming, Fu, Ruibo·2022·Proceedings of the 1st International Workshop on Deepfake Detection for Audio Multimedia
DOI 10.1145/3552466.3556525
Open this paper in CoChat — free accountOpen PDF

Abstract

\beginabstract Many effective attempts have been made for fake audio detection. However, they can only provide detection results but no countermeasures to curb this harm. For many related practical applications, what model or algorithm generated the fake audio also is needed. Therefore, We propose a new problem for detecting vocoder fingerprints of fake audio. Experiments are conducted on the datasets synthesized by eight state-of-the-art vocoders. We have preliminarily explored the features and model architectures. The t-SNE visualization shows that different vocoders generate distinct vocoder fingerprints. \endabstract

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1145/3552466.3556525

Cites

8 of 8 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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?

  • Audio deepfakes exhibit distinct artifact patterns, such as spectral anomalies, phase coherence irregularities, and prosodic inconsistencies, which require specialized detection architectures.

    “Similarly, audio DF exhibit entirely distinct artifact patterns, such as spectral anomalies, phase coherence irregularities, and prosodic inconsistencies, which require specialized detection architectures [13, 14].”

    From Deepfake detection across image, video, and audio: a comprehensive survey with empirical evaluation of generalization and robustness (2026) · cites this paper as [14]

    The Introduction describes distinguishable fingerprints left by different vocoders, and §5.5 identifies a high-frequency spectrogram difference in Griffin-Lim audio. Section 3 describes a detection architecture using acoustic features and a ResNet. The paper does not identify phase-coherence irregularities or prosodic inconsistencies as artifacts; its reference to LPCNet’s “initial phase” concerns the start of the audio, not phase coherence. Its findings concern audio from eight vocoders, rather than audio deepfakes generally. (LLM, conf=0.87)

    Experimental — accuracy not yet measured

  • The spectrograms of original audio have more natural and consistent high-frequency components and more natural harmonic structures compared to synthetic audio.

    “Particularly, these methods have explored that the spectrograms of original audio have more natural and consistent high-frequency components and more natural harmonic structures compared to synthetic audio [13, 14].”

    From Deepfake detection across image, video, and audio: a comprehensive survey with empirical evaluation of generalization and robustness (2026) · cites this paper as [14]

    Section 5.3 says that audio synthesized by eight vocoders differs more in the high frequencies, and Figure 3 compares spectrograms of fake audio from those vocoders. Neither passage compares original with synthetic audio or describes the naturalness or consistency of high-frequency components or harmonic structures. Searches for those comparisons found no supporting passage in our copy. (LLM, conf=0.72)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 32

  1. Heiga Ze, Andrew Senior, and Mike Schuster. Statistical parametric speech synthesis using deep neural networks. In 2013 ieee international conference on acoustics, speech and signal processing, pages 7962-7966. IEEE, 2013.
  2. Yao Qian, Yuchen Fan, Wenping Hu, and Frank K Soong. On the training aspects of deep neural network (dnn) for parametric tts synthesis. In 2014 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 3829-3833. IEEE, 2014.
  3. Heiga Zen and Haşim Sak. Unidirectional long short-term memory recurrent neural network with recurrent output layer for low-latency speech synthesis. In 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 4470-4474. IEEE, 2015.
  4. Wenfu Wang, Shuang Xu, Bo Xu, et al. First step towards end-to-end parametric tts synthesis: Generating spectral parameters with neural attention. In Interspeech, pages 2243-2247, 2016.
  5. Hao Li, Yongguo Kang, and Zhenyu Wang. Emphasis: An emotional phoneme- based acoustic model for speech synthesis system. arXiv preprint arXiv:1806.09276, 2018. · arXiv
  6. Aaron van den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016. · arXiv
  7. Laurens Van der Maaten and Geoffrey Hinton. Visualizing data using t-sne. Journal of machine learning research, 9(11), 2008.
  8. Xin Wang and Junichi Yamagishi. Investigating self-supervised front ends for speech spoofing countermeasures. arXiv preprint arXiv:2111.07725, 2021. · arXiv
  9. Jiangyan Yi, Ye Bai, Jianhua Tao, Zhengkun Tian, Chenglong Wang, Tao Wang, and Ruibo Fu. Half-truth: A partially fake audio detection dataset. arXiv preprint arXiv:2104.03617, 2021. · arXiv
  10. Hemlata Tak, Jose Patino, Massimiliano Todisco, Andreas Nautsch, Nicholas Evans, and Anthony Larcher. End-to-end anti-spoofing with rawnet2. In ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6369-6373. IEEE, 2021.
  11. Haoxin Ma, Jiangyan Yi, Jianhua Tao, Ye Bai, Zhengkun Tian, and Cheng- long Wang. Continual learning for fake audio detection. arXiv preprint arXiv:2104.07286, 2021. · arXiv
  12. Massimiliano Todisco, Héctor Delgado, and Nicholas WD Evans. A new feature for automatic speaker verification anti-spoofing: Constant q cepstral coefficients. In Odyssey, volume 2016, pages 283-290, 2016.
  13. Steven Davis and Paul Mermelstein. Comparison of parametric representations for monosyllabic word recognition in continuously spoken sentences. IEEE transactions on acoustics, speech, and signal processing, 28(4):357-366, 1980.
  14. A Lawson, Pavel Vabishchevich, M Huggins, P Ardis, Brandon Battles, and A Stauffer. Survey and evaluation of acoustic features for speaker recognition. In 2011 IEEE international conference on acoustics, speech and signal processing (ICASSP), pages 5444-5447. IEEE, 2011.
  15. Hessam Bagherinezhad, Mohammad Rastegari, and Ali Farhadi. Lcnn: Lookup- based convolutional neural network. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7120-7129, 2017.
  16. Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation networks. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 7132-7141, 2018.
  17. David Snyder, Daniel Garcia-Romero, Gregory Sell, Daniel Povey, and Sanjeev Khudanpur. X-vectors: Robust dnn embeddings for speaker recognition. In 2018 IEEE international conference on acoustics, speech and signal processing (ICASSP), pages 5329-5333. IEEE, 2018.
  18. Hideki Kawahara. Straight, exploitation of the other aspect of vocoder: Per- ceptually isomorphic decomposition of speech sounds. Acoustical science and technology, 27(6):349-353, 2006.
  19. Masanori Morise, Fumiya Yokomori, and Kenji Ozawa. World: a vocoder-based high-quality speech synthesis system for real-time applications. IEICE TRANS- ACTIONS on Information and Systems, 99(7):1877-1884, 2016.
  20. Jean-Marc Valin and Jan Skoglund. Lpcnet: Improving neural speech synthesis through linear prediction. In ICASSP 2019-2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 5891-5895. IEEE, 2019.
  21. Aaron Oord, Yazhe Li, Igor Babuschkin, Karen Simonyan, Oriol Vinyals, Koray Kavukcuoglu, George Driessche, Edward Lockhart, Luis Cobo, Florian Stimberg, et al. Parallel wavenet: Fast high-fidelity speech synthesis. In International conference on machine learning, pages 3918-3926. PMLR, 2018.
  22. Nal Kalchbrenner, Erich Elsen, Karen Simonyan, Seb Noury, Norman Casagrande, Edward Lockhart, Florian Stimberg, Aaron Oord, Sander Dieleman, and Koray Kavukcuoglu. Efficient neural audio synthesis. In International Conference on Machine Learning, pages 2410-2419. PMLR, 2018.
  23. Ryuichi Yamamoto, Eunwoo Song, and Jae-Min Kim. Parallel wavegan: A fast waveform generation model based on generative adversarial networks with multi- resolution spectrogram. In ICASSP 2020-2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6199-6203. IEEE, 2020.
  24. Jungil Kong, Jaehyeon Kim, and Jaekyoung Bae. Hifi-gan: Generative adversarial networks for efficient and high fidelity speech synthesis. Advances in Neural Information Processing Systems, 33:17022-17033, 2020.
  25. Geng Yang, Shan Yang, Kai Liu, Peng Fang, Wei Chen, and Lei Xie. Multi-band melgan: Faster waveform generation for high-quality text-to-speech. In 2021 IEEE Spoken Language Technology Workshop (SLT), pages 492-498. IEEE, 2021.
  26. Ahmed Mustafa, Nicola Pia, and Guillaume Fuchs. Stylemelgan: An efficient high- fidelity adversarial vocoder with temporal adaptive normalization. In ICASSP 2021-2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pages 6034-6038. IEEE, 2021.
  27. Daniel Griffin and Jae Lim. Signal estimation from modified short-time fourier transform. IEEE Transactions on acoustics, speech, and signal processing, 32(2):236- 243, 1984.
  28. Simone Milani, Marco Fontani, Paolo Bestagini, Mauro Barni, Alessandro Piva, Marco Tagliasacchi, and Stefano Tubaro. An overview on video forensics. APSIPA Transactions on Signal and Information Processing, 1, 2012.
  29. Matthew C Stamm, W Sabrina Lin, and KJ Ray Liu. Temporal forensics and anti-forensics for motion compensated video. IEEE Transactions on Information Forensics and Security, 7(4):1315-1329, 2012.
  30. Ning Yu, Larry S Davis, and Mario Fritz. Attributing fake images to gans: Learning and analyzing gan fingerprints. In Proceedings of the IEEE/CVF international conference on computer vision, pages 7556-7566, 2019.
  31. David Snyder, Daniel Garcia-Romero, Alan McCree, Gregory Sell, Daniel Povey, and Sanjeev Khudanpur. Spoken language recognition using x-vectors. In Odyssey, pages 105-111, 2018.
  32. Galina Lavrentyeva, Sergey Novoselov, Andzhukaev Tseren, Marina Volkova, Artem Gorlanov, and Alexandr Kozlov. Stc antispoofing systems for the asvspoof2019 challenge. arXiv preprint arXiv:1904.05576, 2019. · arXiv

Cited by

Sorted by citations

Authors
Yan, X., Yi, J., Tao, J., Wang, C., Ma, H., Wang, T., Wang, S., & Fu, R.
Affiliations
Chinese Academy of Sciences (CN)
ORCID
Venue
Proceedings of the 1st International Workshop on Deepfake Detection for Audio Multimedia
Publisher
ACM
Published
2022-10-10
Pages
61-68
DOI
10.1145/3552466.3556525
OpenAlex
W4292957223
Cited by
30
References
8
Citations by year
2026: 7 · 2025: 13 · 2024: 6 · 2023: 4
FWCI
3.18
Type
conference paper
Language
en
Open access
Licence
Locations
  • arXiv (Cornell University) (submittedVersion)
  • arXiv (Cornell University)
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
deepfake audio, synthetic audio, t-SNE visualization

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