The DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index.
Are Recent Deepfake Speech Generators Detectable?
Open accessAbstract
Deep learning methods can generate high-quality synthetic speech which is perceptually indistinguishable from real human speech. Synthetic speech can be maliciously used for fraud. Synthetic speech detection methods have been proposed which perform well on ASVspoof2019 and ASVspoof2021 Datasets. These datasets consist of synthetic speech from conventional neural network speech generators. Recently, many voice cloning methods have been proposed which use diffusion models and generative adversarial networks for high-quality speech synthesis. In this work, we present a new synthetic speech dataset containing 25,000 synthetic speech signals for 11 distinct speakers, with a total duration of 52 hours. We have developed this dataset using 5 recent diffusion model-based synthetic speech generators. These generators can clone a speaker's voice from text using only a few minutes of their real speech. We evaluate 6 of the best synthetic speech detectors that work well on the ASVspoof2019 Dataset on this new dataset, and demonstrate their performance using Equal Error Rate (EER).
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
“Purdue Speech Dataset [45] includes 25,000 synthetic speech tracks generated using five advanced diffusion model-based voice cloning methods (ProDiff, DiffGAN-TTS, ElevenLabs, UnitSpeech, and XTTS).”
From Deepfake Media Forensics: Status and Future Challenges (2025) · cites this paper as [45]
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 3×
“These features are input to a ResNet network [16] which classifies the input speech as real or synthetic.”
BackgroundMethodGoing deeper with convolutions
Szegedy, C., Liu, W., Jia, Y., Sermanet, P., Reed, S., Anguelov, D., Erhan, D., Vanhoucke, V., & Rabinovich, A. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298594 · Cited by 48,562
Learning representations by back-propagating errors
Rumelhart, D. E., Hinton, G. E., & Williams, R. J. (1986) · Nature · doi:10.1038/323533a0 · Cited by 31,892 · cited 1×
“[10] used a convolutional Recurrent Neural Network (RNN) [33] for synthetic speech detection.”
Method“Librispeech: an ASR corpus based on public domain audio books,”
Panayotov, V., Chen, G., Povey, D., & Khudanpur, S. (2015) · 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2015.7178964 · Cited by 6,072 · cited 2×
“With an aim to test the performance of all three zero-shot voice cloning methods (Eleven-Labs, UnitSpeech, and YourTTS) on multiple speakers, we chose 10 speakers, 5 female and 5 male from the LibriSpeech Dataset [30].”
MethodAudio Set: An ontology and human-labeled dataset for audio events
Gemmeke, J. F., Ellis, D. P. W., Freedman, D., Jansen, A., Lawrence, W., Moore, R. C., Plakal, M., & Ritter, M. (2017) · 2017 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp.2017.7952261 · Cited by 3,134 · cited 3×
“The transformer is pretrained on a large real speech dataset known as AudioSet [14] for audio recognition [24].”
BackgroundMethodInfluentialRay, P. P. (2023) · Internet of Things and Cyber-Physical Systems · doi:10.1016/j.iotcps.2023.04.003 · Cited by 2,463
A Scale for the Measurement of the Psychological Magnitude Pitch
Stevens, S. S., Volkmann, J., & Newman, E. B. (1937) · The Journal of the Acoustical Society of America · doi:10.1121/1.1915893 · Cited by 1,107 · cited 1×
“D representation of speech obtained using Short Time Fourier Transform (STFT), with time on x-axis and frequency in mel scale on y-axis [6, 7, 35].”
MethodHiFi-GAN: Generative Adversarial Networks for Efficient and High Fidelity Speech Synthesis
Kong, J., Kim, J., & Bae, J. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2010.05646 · Cited by 722 · cited 3×
“In [19], a modified diffusion model is used to generate mel-spectrogram of the speech of the target speaker, which is input to the HiFi-GAN neural vocoder [23] used for speech synthesis.”
MethodConstant Q Cepstral Coefficients: A Spoofing Countermeasure for Automatic Speaker Verification
Todisco, M., Delgado, H., & Evans, N. (2017) · Computer Speech & Language · doi:10.1016/j.csl.2017.01.001 · Cited by 424 · cited 1×
“Speech feature-based methods use selected features derived from speech, for example, Mel-Frequency Cepstrum Coefficients (MFCC) [4], Constant Q Cepstral Coefficients (CQCC) [38], and Linear Frequency Cepstral Coefficients (LFCC) [26] to detect synthetic speech.”
MethodASVspoof 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
Research works printed but not linked: 10
- 10.21437/Interspeech.2019-3174 · DOI · Paper page
- Karlheinz Brandenburg. 1999. MP3 and AAC Explained. Proceedings of the International Conference of Audio Engineering Society: High-Quality Audio Coding (September 1999). Signa, Italy.
- Andrew Brock, Jeff Donahue, and Karen Simonyan. 2019. Large Scale GAN Training for High Fidelity Natural Image Synthesis. Proceedings of the International Conference on Learning Representations (May 2019), 35 pages. https: //openreview.net/pdf?id=B1xsqj09Fm New Orleans, Louisiana, USA.
- 10.1007/978-981-99-1912-3_21 · DOI
- Jonathan Ho, Ajay Jain, and Pieter Abbeel. 2020. Denoising Diffusion Probabilistic Models. Advances in Neural Information Processing Systems 33 (December 2020), 6840--6851. https://proceedings.neurips.cc/paper_files/paper/2020/file/ 4c5bcfec8584af0d967f1ab10179ca4b-Paper.pdf
- 10.21437/Interspeech.2023-2326 · DOI
- 10.21437/Interspeech.2022-227 · DOI · Paper page
- Alec Radford, Karthik Narasimhan, Tim Salimans, and Ilya Sutskever. 2018. Improving Language Understanding by Generative Pre-training. (June 2018), 12 pages. OpenAI.
- Zhisheng Xiao, Karsten Kreis, and Arash Vahdat. 2022. Tackling the Generative Learning Trilemma with Denoising Diffusion GANs. Proceedings of the International Conference on Learning Representations (April 2022), 28 pages. https://openreview.net/pdf?id=JprM0p-q0Co
- 10.7488/ds/2555 · DOI
Cited by
Deepfake Media Forensics: Status and Future Challenges
Amerini, I., Barni, M., Battiato, S., Bestagini, P., Boato, G., Bruni, V., Caldelli, R., De Natale, F., De Nicola, R., Guarnera, L., et al. (2025) · Journal of Imaging · doi:10.3390/jimaging11030073 · Cited by 113
“Purdue Speech Dataset [45] includes 25,000 synthetic speech tracks generated us-ing five advanced diffusion model-based voice cloning methods (ProDiff, DiffGAN-TTS, ElevenLabs, UnitSpeech, and XTTS).”
Self-citationLeveraging Mixture of Experts for Improved Speech Deepfake Detection
Negroni, V., Salvi, D., Mezza, A. I., Bestagini, P., & Tubaro, S. (2025) · ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · doi:10.1109/icassp49660.2025.10890398 · Cited by 14
“Purdue speech dataset [1] ( D PUR ).”
BackgroundSelf-citationDiffSSD: A Diffusion-Based Dataset For Speech Forensics
Bhagtani, K., Yadav, A. K. S., Bestagini, P., & Delp, E. J. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889450 · Cited by 18
Self-citationPhoneme-Level Analysis for Person-of-Interest Speech Deepfake Detection
Salvi, D., Negroni, V., Mandelli, S., Bestagini, P., & Tubaro, S. (2025) · IEEE International Conference on Computer Vision Workshops · doi:10.1109/iccvw69036.2025.00169 · Cited by 15
“Purdue speech dataset [7] .”
Self-citationAnomaly Detection and Localization for Speech Deepfakes via Feature Pyramid Matching
Coletta, E., Salvi, D., Negroni, V., Leonzio, D. U., & Bestagini, P. (2025) · European Signal Processing Conference (EUSIPCO) · doi:10.23919/eusipco63237.2025.11226156 · Cited by 7
Self-citation
Metadata
- Authors
- Bhagtani, K., Yadav, A. K. S., Bestagini, P., & Delp, E. J.
- Affiliations
- Purdue University West Lafayette (US)
- Politecnico di Milano (IT)
- Venue
- Proceedings of the 2024 ACM Workshop on Information Hiding and Multimedia Security
- Publisher
- ACM
- Published
- 2024-06-24
- Pages
- 277-282
- OpenAlex
- W4399556715
- Cited by
- 9
- References
- 28
- Citations by year
- 2026: 1 · 2025: 8
- FWCI
- 2.60
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- public-domain
- creativecommons.org/licenses/by/4.0/ (vor)
- Locations
- ACM Workshop on Information Hiding and Multimedia Security (IH&MMSec) (publishedVersion, public-domain)
- Funders
- AFRL and DARPA (FA8750-20-2-1004)
- 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 › Computer Vision and Pattern Recognition › Digital Media Forensic Detection
- Keywords
- speech deepfake detection, synthetic speech, voice cloning, diffusion models, generative adversarial networks, ASVspoof, anti-spoofing, speaker verification
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: Are Recent Deepfake Speech Generators Detectable? https://app.cochat.ai/papers/10.1145/3658664.3659658
- The paper
- Bhagtani, K., Yadav, A. K. S., Bestagini, P., & Delp, E. J. (2024). Are Recent Deepfake Speech Generators Detectable? Proceedings of the 2024 ACM Workshop on Information Hiding and Multimedia Security. https://doi.org/10.1145/3658664.3659658
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.