Publication result detail

Deepfake Speech Detection: A Spectrogram Analysis

FIRC, A.; MALINKA, K.; HANÁČEK, P.

Original Title

Deepfake Speech Detection: A Spectrogram Analysis

English Title

Deepfake Speech Detection: A Spectrogram Analysis

Type

Paper in proceedings (conference paper)

Original Abstract

The current voice biometric systems have no natural mechanics to defend against deepfake spoofing attacks. Thus, supporting these systems with a deepfake detection solution is necessary. One of the latest approaches to deepfake speech detection is representing speech as a spectrogram and using it as an input for a deep neural network. This work thus analyzes the feasibility of different spectrograms for deepfake speech detection. We compare types of them regarding their performance, hardware requirements, and speed. We show the majority of the spectrograms are feasible for deepfake detection. However, there is no general, correct answer to selecting the best spectrogram. As we demonstrate, different spectrograms are suitable for different needs.

English abstract

The current voice biometric systems have no natural mechanics to defend against deepfake spoofing attacks. Thus, supporting these systems with a deepfake detection solution is necessary. One of the latest approaches to deepfake speech detection is representing speech as a spectrogram and using it as an input for a deep neural network. This work thus analyzes the feasibility of different spectrograms for deepfake speech detection. We compare types of them regarding their performance, hardware requirements, and speed. We show the majority of the spectrograms are feasible for deepfake detection. However, there is no general, correct answer to selecting the best spectrogram. As we demonstrate, different spectrograms are suitable for different needs.

Keywords

Deepfake, Speech, Image-based, Deepfake Detection, Spectrogram

Key words in English

Deepfake, Speech, Image-based, Deepfake Detection, Spectrogram

Authors

FIRC, A.; MALINKA, K.; HANÁČEK, P.

RIV year

2025

Released

08.04.2024

Publisher

Association for Computing Machinery

Location

Avila

ISBN

979-8-4007-0243-3

Book

Proceedings of the ACM Symposium on Applied Computing

Pages from

1312

Pages to

1320

Pages count

9

URL

https://dl.acm.org/doi/10.1145/3605098.3635911

Full text in the Digital Library

http://hdl.handle.net/11012/252867

BibTex

@inproceedings{BUT188028,
  author="Anton {Firc} and Kamil {Malinka} and Petr {Hanáček}",
  title="Deepfake Speech Detection: A Spectrogram Analysis",
  booktitle="Proceedings of the ACM Symposium on Applied Computing",
  year="2024",
  pages="1312--1320",
  publisher="Association for Computing Machinery",
  address="Avila",
  doi="10.1145/3605098.3635911",
  isbn="979-8-4007-0243-3",
  url="https://dl.acm.org/doi/10.1145/3605098.3635911"
}

Documents

3605098.3635911