Every Breath You Don’t Take: Deepfake Speech Detection Using Breath
Digital Threats: Research and Practice · 30.09.2025 · 6 (3) · 1-18
DOI: 10.1145/3754456 ↗Kurz gefasst
Fachartikel in der Zeitschrift Digital Threats: Research and Practice, online erschienen am 12.09.2025. Verfasst von Seth Layton und 6 weiteren Autor:innen. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 49 Literaturangaben.
Abstract (Originalsprache)
Deepfake speech represents a real and growing threat to systems and society. Many detectors have been created to aid in defense against speech deepfakes. While these detectors implement myriad methodologies, many rely on low-level fragments of the speech generation process. We hypothesize that breath, a higher-level part of speech, is a key component of natural speech and thus improper generation in deepfake speech is a performant discriminator. To evaluate this, we create a breath detector and leverage this against a custom dataset of online news article audio to discriminate between real/deepfake speech. Additionally, we make this custom dataset publicly available to facilitate comparison for future work. Applying our simple breath detector as a deepfake speech discriminator on in-the-wild samples allows for accurate classification (perfect 1.0 AUPRC and 0.0 EER on test data) across 33.6 hours of audio. We compare our model with the state-of-the-art SSL-wav2vec and Codecfake models and show that these complex deep learning model completely either fail to classify the same in-the-wild samples (0.72 AUPRC and 0.89 EER), or substantially lack in the computational and temporal performance compared to our methodology (37 seconds to predict a 1 minute sample with Codecfake vs. 0.3 seconds with our model).
Zitieren
APA
Layton, S., Andrade, T. D., Olszewski, D., Warren, K., Gates, C., Butler, K., & Traynor, P. (2025). Every Breath You Don’t Take: Deepfake Speech Detection Using Breath. Digital Threats: Research and Practice, 6(3), 1-18. https://doi.org/10.1145/3754456
Vancouver
Layton S, Andrade TD, Olszewski D, Warren K, Gates C, Butler K, et al. Every Breath You Don’t Take: Deepfake Speech Detection Using Breath. Digital Threats: Research and Practice. 2025;6(3):1-18. doi:10.1145/3754456
BibTeX
@article{Layton2025,
author = {Layton, Seth and Andrade, Thiago De and Olszewski, Daniel and Warren, Kevin and Gates, Carrie and Butler, Kevin and Traynor, Patrick},
title = {Every Breath You Don’t Take: Deepfake Speech Detection Using Breath},
journal = {Digital Threats: Research and Practice},
year = {2025},
volume = {6},
number = {3},
pages = {1--18},
doi = {10.1145/3754456}
}
EUVIMED