ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Xin, Delgado, Hector, Tak, Hemlata, Jung, Jee-weon, Shim, Hye-jin, Todisco, Massimiliano, Kukanov, Ivan, Liu, Xuechen, Sahidullah, Md, Kinnunen, Tomi, Evans, Nicholas, Lee, Kong Aik, Yamagishi, Junichi
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911991409410048
author Wang, Xin
Delgado, Hector
Tak, Hemlata
Jung, Jee-weon
Shim, Hye-jin
Todisco, Massimiliano
Kukanov, Ivan
Liu, Xuechen
Sahidullah, Md
Kinnunen, Tomi
Evans, Nicholas
Lee, Kong Aik
Yamagishi, Junichi
author_facet Wang, Xin
Delgado, Hector
Tak, Hemlata
Jung, Jee-weon
Shim, Hye-jin
Todisco, Massimiliano
Kukanov, Ivan
Liu, Xuechen
Sahidullah, Md
Kinnunen, Tomi
Evans, Nicholas
Lee, Kong Aik
Yamagishi, Junichi
contents ASVspoof 5 is the fifth edition in a series of challenges that promote the study of speech spoofing and deepfake attacks, and the design of detection solutions. Compared to previous challenges, the ASVspoof 5 database is built from crowdsourced data collected from a vastly greater number of speakers in diverse acoustic conditions. Attacks, also crowdsourced, are generated and tested using surrogate detection models, while adversarial attacks are incorporated for the first time. New metrics support the evaluation of spoofing-robust automatic speaker verification (SASV) as well as stand-alone detection solutions, i.e., countermeasures without ASV. We describe the two challenge tracks, the new database, the evaluation metrics, baselines, and the evaluation platform, and present a summary of the results. Attacks significantly compromise the baseline systems, while submissions bring substantial improvements.
format Preprint
id arxiv_https___arxiv_org_abs_2408_08739
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale
Wang, Xin
Delgado, Hector
Tak, Hemlata
Jung, Jee-weon
Shim, Hye-jin
Todisco, Massimiliano
Kukanov, Ivan
Liu, Xuechen
Sahidullah, Md
Kinnunen, Tomi
Evans, Nicholas
Lee, Kong Aik
Yamagishi, Junichi
Audio and Speech Processing
Artificial Intelligence
Sound
ASVspoof 5 is the fifth edition in a series of challenges that promote the study of speech spoofing and deepfake attacks, and the design of detection solutions. Compared to previous challenges, the ASVspoof 5 database is built from crowdsourced data collected from a vastly greater number of speakers in diverse acoustic conditions. Attacks, also crowdsourced, are generated and tested using surrogate detection models, while adversarial attacks are incorporated for the first time. New metrics support the evaluation of spoofing-robust automatic speaker verification (SASV) as well as stand-alone detection solutions, i.e., countermeasures without ASV. We describe the two challenge tracks, the new database, the evaluation metrics, baselines, and the evaluation platform, and present a summary of the results. Attacks significantly compromise the baseline systems, while submissions bring substantial improvements.
title ASVspoof 5: Crowdsourced Speech Data, Deepfakes, and Adversarial Attacks at Scale
topic Audio and Speech Processing
Artificial Intelligence
Sound
url https://arxiv.org/abs/2408.08739