Warning: Humans cannot reliably detect speech deepfakes
PLOS ONE · 02.08.2023 · 18 (8) · e0285333
DOI: 10.1371/journal.pone.0285333 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS ONE, online erschienen am 02.08.2023. Verfasst von Kimberly T Mai und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Security and Crime Science, University College London, London, United Kingdom. Hauptthemen laut MEDLINE: Speech, Speech Perception. Die Arbeit wurde bisher 105-mal zitiert (Crossref) und nennt 51 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Engineering and Physical Sciences Research Council, Dawes Centre for Future Crime.
Abstract (Originalsprache)
Speech deepfakes are artificial voices generated by machine learning models. Previous literature has highlighted deepfakes as one of the biggest security threats arising from progress in artificial intelligence due to their potential for misuse. However, studies investigating human detection capabilities are limited. We presented genuine and deepfake audio to n = 529 individuals and asked them to identify the deepfakes. We ran our experiments in English and Mandarin to understand if language affects detection performance and decision-making rationale. We found that detection capability is unreliable. Listeners only correctly spotted the deepfakes 73% of the time, and there was no difference in detectability between the two languages. Increasing listener awareness by providing examples of speech deepfakes only improves results slightly. As speech synthesis algorithms improve and become more realistic, we can expect the detection task to become harder. The difficulty of detecting speech deepfakes confirms their potential for misuse and signals that defenses against this threat are needed.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- Department of Security and Crime Science, University College London, London, United Kingdom.
- Department of Computer Science, University College London, London, United Kingdom.
ORCID-Profile: Kimberly T Mai, Sergi Bray, Lewis D Griffin
Förderung
- Engineering and Physical Sciences Research Council
- Dawes Centre for Future Crime
Zitieren
APA
Mai, K. T., Bray, S., Davies, T., & Griffin, L. D. (2023). Warning: Humans cannot reliably detect speech deepfakes. PLOS ONE, 18(8), e0285333. https://doi.org/10.1371/journal.pone.0285333
Vancouver
Mai KT, Bray S, Davies T, Griffin LD. Warning: Humans cannot reliably detect speech deepfakes. PLOS ONE. 2023;18(8):e0285333. doi:10.1371/journal.pone.0285333
BibTeX
@article{Mai2023,
author = {Mai, Kimberly T and Bray, Sergi and Davies, Toby and Griffin, Lewis D},
title = {Warning: Humans cannot reliably detect speech deepfakes},
journal = {PLOS ONE},
year = {2023},
volume = {18},
number = {8},
pages = {e0285333},
doi = {10.1371/journal.pone.0285333}
}
EUVIMED