Publication Details

Assessing the Human Ability to Recognize Synthetic Speech in Ordinary Conversation

PRUDKÝ, D.; FIRC, A.; MALINKA, K. Assessing the Human Ability to Recognize Synthetic Speech in Ordinary Conversation. In 2023 International Conference of the Biometrics Special Interest Group (BIOSIG). Proceedings of the 22nd International Conference of the Biometrics Special Interest Group. Darmstadt: GI - Group for computer science, 2023. p. 1-5. ISBN: 978-3-88579-733-3.
Czech title
Hodnocení lidské schopnosti rozpoznat syntetickou řeč v běžné konverzaci
Type
conference paper
Language
English
Authors
Prudký Daniel, Bc.
Firc Anton, Ing. (DITS)
Malinka Kamil, Mgr., Ph.D. (DITS)
URL
Keywords

deepfake, synthetic speech, artificial intelligence, cybersecurity, deepfake
detection

Abstract

This work assesses the human ability to recognize synthetic speech (deepfake).
This paper describes an experiment in which we communicated with respondents
using voice messages. We presented the respondents with a cover story about
testing the user-friendliness of voice messages while secretly sending them
a pre-prepared deep fake recording during the conversation. We examined their
reactions, knowledge of deepfakes, or how many could correctly identify which
message was deepfake. The results show that none of the respondents reacted in
any way to the fraudulent deepfake message, and only one retrospectively admitted
to noticing something specific. On the other hand, a voicemail message that
contained a deepfake was correctly identified by 83.9% of respondents after
revealing the nature of the experiment. Thus, the results show that although the
deepfake recording was clearly identifiable among others, no one reacted to it.
In summary, we show that the human ability to recognize voice deepfakes is not at
a level we can trust. It is very difficult for people to distinguish between real
and fake voices, especially if they do not expect them.

Published
2023
Pages
1–5
Proceedings
2023 International Conference of the Biometrics Special Interest Group (BIOSIG)
Series
Proceedings of the 22nd International Conference of the Biometrics Special Interest Group
Conference
22nd International Conference of the Biometrics Special Interest Group, Darmstadt, DE
ISBN
978-3-88579-733-3
Publisher
GI - Group for computer science
Place
Darmstadt
DOI
EID Scopus
BibTeX
@inproceedings{BUT185186,
  author="Daniel {Prudký} and Anton {Firc} and Kamil {Malinka}",
  title="Assessing the Human Ability to Recognize Synthetic Speech in Ordinary Conversation",
  booktitle="2023 International Conference of the Biometrics Special Interest Group (BIOSIG)",
  year="2023",
  series="Proceedings of the 22nd International Conference of the Biometrics Special Interest Group",
  pages="1--5",
  publisher="GI - Group for computer science",
  address="Darmstadt",
  doi="10.1109/BIOSIG58226.2023.10346006",
  isbn="978-3-88579-733-3",
  url="https://ieeexplore.ieee.org/document/10346006"
}
Back to top