Detail výsledku

Comprehensive Multiparametric Analysis of Human Deepfake Speech Recognition

MALINKA, K.; FIRC, A.; ŠALKO, M.; PRUDKÝ, D.; RADAČOVSKÁ, K.; HANÁČEK, P. Comprehensive Multiparametric Analysis of Human Deepfake Speech Recognition. EURASIP Journal on Image and Video Processing, 2024, vol. 2024, no. 24, p. 1-25. ISSN: 1687-5281.
Typ
článek v časopise
Jazyk
anglicky
Autoři
Malinka Kamil, doc. Mgr., Ph.D., UITS (FIT)
Firc Anton, Ing., UITS (FIT)
Šalko Milan, Ing., UITS (FIT)
Prudký Daniel, Ing.
Radačovská Karolína, Bc., UPGM (FIT)
Hanáček Petr, doc. Dr. Ing., UITS (FIT)
Abstrakt

In this paper, we undertake a novel two-pronged investigation into the human recognition of deepfake speech, addressing critical gaps in existing research. First, we pioneer an evaluation of the impact of prior information on deepfake recognition, setting our work apart by simulating real-world attack scenarios where individuals are not informed in advance of deepfake exposure. This approach simulates the unpredictability of real-world deepfake attacks, providing unprecedented insights into human vulnerability under realistic conditions. Second, we introduce a novel metric to evaluate the quality of deepfake audio. This metric facilitates a deeper exploration into how the quality of deepfake speech influences human detection accuracy. By examining both the effect of prior knowledge about deepfakes and the role of deepfake speech quality, our research reveals the importance of these factors, contributes to understanding human vulnerability to deepfakes, and suggests measures to enhance human detection skills.

Klíčová slova

Deepfake, Synthetic speech, Deepfake detection, Human perception, Speech quality, Cybersecurity

URL
Rok
2024
Strany
1–25
Časopis
EURASIP Journal on Image and Video Processing, roč. 2024, č. 24, ISSN 1687-5281
Vydavatel
Springer Nature
DOI
UT WoS
001302501400001
EID Scopus
BibTeX
@article{BUT189344,
  author="Kamil {Malinka} and Anton {Firc} and Milan {Šalko} and Daniel {Prudký} and Karolína {Radačovská} and Petr {Hanáček}",
  title="Comprehensive Multiparametric Analysis of Human Deepfake Speech Recognition",
  journal="EURASIP Journal on Image and Video Processing",
  year="2024",
  volume="2024",
  number="24",
  pages="1--25",
  doi="10.1186/s13640-024-00641-4",
  issn="1687-5176",
  url="https://jivp-eurasipjournals.springeropen.com/articles/10.1186/s13640-024-00641-4"
}
Soubory
Projekty
Reliable, Secure, and Intelligent Computer Systems, VUT, Vnitřní projekty VUT, FIT-S-23-8151, zahájení: 2023-03-01, ukončení: 2026-02-28, řešení
Výzkumné skupiny
Pracoviště
Nahoru