The DOI matches its Crossref record.
Towards generalisable and calibrated audio deepfake detection with self-supervised representations
Open accessAbstract
Generalisation—the ability of a model to perform well on unseen data—is crucial for building reliable deepfake detectors. However, recent studies have shown that the current audio deepfake models fall short of this desideratum. In this work we investigate the potential of pretrained self-supervised representations in building general and calibrated audio deepfake detection models. We show that large frozen representations coupled with a simple logistic regression classifier are extremely effective in achieving strong generalisation capabilities: compared to the RawNet2 model, this approach reduces the equal error rate from 30.9% to 8.8% on a benchmark of eight deepfake datasets, while learning less than 2k parameters. Moreover, the proposed method produces considerably more reliable predictions compared to previous approaches making it more suitable for realistic use.
Overview
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.
Integrity signals
Registry
Cites
Statistics
Verification
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.
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“WavLM and HuBERT representations have also been employed [1,4,5,8], but their results seem to trail those of wav2vec.”
From WavLM model ensemble for audio deepfake detection (2024) · cites this paper as [8]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“As shown above, large SSL models have shown very good performance over the deepfake detection task.”
From WavLM model ensemble for audio deepfake detection (2024) · cites this paper as [8]
The cited source says what the claim attributes to it.
In the citing paper
“This is also the case for our task of interest, audio deepfake detection, where methods based on self-supervised representations provide a lighter [7] and more robust [8] alternative compared to the previous generation of approaches, such as ResNet on linear frequency cepstral coefficients (LFCC) [14] or RawNet [15].”
From WavLM model ensemble for audio deepfake detection (2024) · cites this paper as [8]
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cited by
Sorted by citations
WavLM model ensemble for audio deepfake detection
Combei, D., Stan, A., Oneata, D., & Cucu, H. (2024) · The Automatic Speaker Verification Spoofing Countermeasures Workshop (ASVspoof 2024) · doi:10.21437/asvspoof.2024-25 · Cited by 15
“Backends range from simple linear models [7,8] to more complex pooling mechanisms [1, 2, 5, 6].”
BackgroundMethodInfluentialSelf-citation
Metadata
- Authors
- Pascu, O., Stan, A., Oneata, D., Oneata, E., & Cucu, H.
- Venue
- Interspeech 2024
- Publisher
- ISCA
- Published
- 2024-09-01
- Pages
- 4828-4832
- ISSN
- 2308-457X (ISSN-L)
- OpenAlex
- W4402111467
- Cited by
- 16
- References
- 0
- Citations by year
- 2026: 4 · 2025: 12
- FWCI
- 4.62
- Type
- conference paper
- Language
- en
- Open access
- green
- doi.org
- Licence
- cc-by
- Locations
- Interspeech (USB) (publishedVersion)
- Zenodo (CERN European Organization for Nuclear Research) (cc-by)
- Funders
- European Commission (101070190, 101120237)
- Unitatea Executiva pentru Finantarea Invatamantului Superior, a Cercetarii, Dezvoltarii si Inovarii
- Colegiul Consultativ pentru Cercetare-Dezvoltare şi Inovare
- 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
- audio deepfake detection, self-supervised representations, generalization, calibration, logistic regression classifier, equal error rate
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
