ASVspoof 5: Evaluation of Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced Speech

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Wang, Xin, Delgado, Héctor, Evans, Nicholas, Liu, Xuechen, Kinnunen, Tomi, Tak, Hemlata, Lee, Kong Aik, Kukanov, Ivan, Sahidullah, Md, Todisco, Massimiliano, Yamagishi, Junichi
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866911586237546496
author Wang, Xin
Delgado, Héctor
Evans, Nicholas
Liu, Xuechen
Kinnunen, Tomi
Tak, Hemlata
Lee, Kong Aik
Kukanov, Ivan
Sahidullah, Md
Todisco, Massimiliano
Yamagishi, Junichi
author_facet Wang, Xin
Delgado, Héctor
Evans, Nicholas
Liu, Xuechen
Kinnunen, Tomi
Tak, Hemlata
Lee, Kong Aik
Kukanov, Ivan
Sahidullah, Md
Todisco, Massimiliano
Yamagishi, Junichi
contents ASVspoof 5 is the fifth edition in a series of challenges which promote the study of speech spoofing and deepfake detection solutions. A significant change from previous challenge editions is a new crowdsourced database collected from a substantially greater number of speakers under diverse recording conditions, and a mix of cutting-edge and legacy generative speech technology. With the new database described elsewhere, we provide in this paper an overview of the ASVspoof 5 challenge results for the submissions of 53 participating teams. While many solutions perform well, performance degrades under adversarial attacks and the application of neural encoding/compression schemes. Together with a review of post-challenge results, we also report a study of calibration in addition to other principal challenges and outline a road-map for the future of ASVspoof.
format Preprint
id arxiv_https___arxiv_org_abs_2601_03944
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ASVspoof 5: Evaluation of Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced Speech
Wang, Xin
Delgado, Héctor
Evans, Nicholas
Liu, Xuechen
Kinnunen, Tomi
Tak, Hemlata
Lee, Kong Aik
Kukanov, Ivan
Sahidullah, Md
Todisco, Massimiliano
Yamagishi, Junichi
Signal Processing
Sound
ASVspoof 5 is the fifth edition in a series of challenges which promote the study of speech spoofing and deepfake detection solutions. A significant change from previous challenge editions is a new crowdsourced database collected from a substantially greater number of speakers under diverse recording conditions, and a mix of cutting-edge and legacy generative speech technology. With the new database described elsewhere, we provide in this paper an overview of the ASVspoof 5 challenge results for the submissions of 53 participating teams. While many solutions perform well, performance degrades under adversarial attacks and the application of neural encoding/compression schemes. Together with a review of post-challenge results, we also report a study of calibration in addition to other principal challenges and outline a road-map for the future of ASVspoof.
title ASVspoof 5: Evaluation of Spoofing, Deepfake, and Adversarial Attack Detection Using Crowdsourced Speech
topic Signal Processing
Sound
url https://arxiv.org/abs/2601.03944