Publication Details

Diffuse or Confuse: A Diffusion Deepfake Speech Dataset

FIRC, A.; MALINKA, K.; HANÁČEK, P. Diffuse or Confuse: A Diffusion Deepfake Speech Dataset. In 2024 International Conference of the Biometrics Special Interest Group (BIOSIG). Darmstadt: GI - Group for computer science, 2024. p. 1-7. ISBN: 978-3-88579-749-4.
Czech title
Difuze nebo zmatení: Difuzní datová sada pro deepfake řeč
Type
conference paper
Language
English
Authors
URL
Keywords

deepfakes, deepfake speech, dataset, diffusion, detection

Abstract

Advancements in artificial intelligence and machine learning have significantly
improved synthetic speech generation. This paper explores diffusion models,
a novel method for creating realistic synthetic speech. We create a diffusion
dataset using available tools and pretrained models. Additionally, this study
assesses the quality of diffusion-generated deepfakes versus non-diffusion ones
and their potential threat to current deepfake detection systems. Findings
indicate that the detection of diffusion-based deepfakes is generally comparable
to non-diffusion deepfakes, with some variability based on detector architecture.
Re-vocoding with diffusion vocoders shows minimal impact, and the overall speech
quality is comparable to non-diffusion methods.

Published
2024
Pages
1–7
Proceedings
2024 International Conference of the Biometrics Special Interest Group (BIOSIG)
Conference
23rd International Conference of the Biometrics Special Interest Group, Darmstadt, DE
ISBN
978-3-88579-749-4
Publisher
GI - Group for computer science
Place
Darmstadt
DOI
BibTeX
@inproceedings{BUT189345,
  author="Anton {Firc} and Kamil {Malinka} and Petr {Hanáček}",
  title="Diffuse or Confuse: A Diffusion Deepfake Speech Dataset",
  booktitle="2024 International Conference of the Biometrics Special Interest Group (BIOSIG)",
  year="2024",
  pages="1--7",
  publisher="GI - Group for computer science",
  address="Darmstadt",
  doi="10.1109/BIOSIG61931.2024.10786752",
  isbn="978-3-88579-749-4",
  url="https://ieeexplore.ieee.org/document/10786752"
}
Back to top