Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals

Tomi Kinnunen; Hector Delgado; Nicholas Evans; Kong Aik Lee; Ville Vestman; Andreas Nautsch; Massimiliano Todisco; Xin Wang; Md Sahidullah; Junichi Yamagishi; Douglas A. Reynolds

doi:10.1109/TASLP.2020.3009494

Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals

Tomi Kinnunen, Hector Delgado, Nicholas Evans, Kong Aik Lee, Ville Vestman, Andreas Nautsch, Massimiliano Todisco, Xin Wang, Md Sahidullah, Junichi Yamagishi, Douglas A. Reynolds

Research output: Journal article publication › Journal article › Academic research › peer-review

81 Citations (Scopus)

Abstract

Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and the impact of spoofing and CMs upon ASV and its use as a primary metric in traditional ASV research has long been abandoned in favour of risk-based approaches to assessment. This paper presents several new extensions to the tandem detection cost function (t-DCF), a recent risk-based approach to assess the reliability of spoofing CMs deployed in tandem with an ASV system. Extensions include a simplified version of the t-DCF with fewer parameters, an analysis of a special case for a fixed ASV system, simulations which give original insights into its interpretation and new analyses using the ASVspoof 2019 database. It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.

Original language	English
Article number	9143410
Pages (from-to)	2195-2210
Number of pages	16
Journal	IEEE/ACM Transactions on Audio Speech and Language Processing
Volume	28
DOIs	https://doi.org/10.1109/TASLP.2020.3009494
Publication status	Published - Jul 2020
Externally published	Yes

Keywords

Automatic speaker verification (ASV)
detect- ion cost function
presentation attack detection
spoofing counter- measures

ASJC Scopus subject areas

Computer Science (miscellaneous)
Acoustics and Ultrasonics
Computational Mathematics
Electrical and Electronic Engineering

More information

10.1109/TASLP.2020.3009494

Cite this

Kinnunen, T., Delgado, H., Evans, N., Lee, K. A., Vestman, V., Nautsch, A., Todisco, M., Wang, X., Sahidullah, M., Yamagishi, J., & Reynolds, D. A. (2020). Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals. IEEE/ACM Transactions on Audio Speech and Language Processing, 28, 2195-2210. Article 9143410. https://doi.org/10.1109/TASLP.2020.3009494

@article{367f6ea9a9ac45efb041f0fb04394ba8,

title = "Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals",

abstract = "Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and the impact of spoofing and CMs upon ASV and its use as a primary metric in traditional ASV research has long been abandoned in favour of risk-based approaches to assessment. This paper presents several new extensions to the tandem detection cost function (t-DCF), a recent risk-based approach to assess the reliability of spoofing CMs deployed in tandem with an ASV system. Extensions include a simplified version of the t-DCF with fewer parameters, an analysis of a special case for a fixed ASV system, simulations which give original insights into its interpretation and new analyses using the ASVspoof 2019 database. It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.",

keywords = "Automatic speaker verification (ASV), detect- ion cost function, presentation attack detection, spoofing counter- measures",

author = "Tomi Kinnunen and Hector Delgado and Nicholas Evans and Lee, {Kong Aik} and Ville Vestman and Andreas Nautsch and Massimiliano Todisco and Xin Wang and Md Sahidullah and Junichi Yamagishi and Reynolds, {Douglas A.}",

note = "Publisher Copyright: {\textcopyright} 2014 IEEE.",

year = "2020",

month = jul,

doi = "10.1109/TASLP.2020.3009494",

language = "English",

volume = "28",

pages = "2195--2210",

journal = "IEEE/ACM Transactions on Audio Speech and Language Processing",

issn = "2329-9290",

publisher = "IEEE Advancing Technology for Humanity",

}

Kinnunen, T, Delgado, H, Evans, N, Lee, KA, Vestman, V, Nautsch, A, Todisco, M, Wang, X, Sahidullah, M, Yamagishi, J & Reynolds, DA 2020, 'Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals', IEEE/ACM Transactions on Audio Speech and Language Processing, vol. 28, 9143410, pp. 2195-2210. https://doi.org/10.1109/TASLP.2020.3009494

TY - JOUR

T1 - Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals

AU - Kinnunen, Tomi

AU - Delgado, Hector

AU - Evans, Nicholas

AU - Lee, Kong Aik

AU - Vestman, Ville

AU - Nautsch, Andreas

AU - Todisco, Massimiliano

AU - Wang, Xin

AU - Sahidullah, Md

AU - Yamagishi, Junichi

AU - Reynolds, Douglas A.

PY - 2020/7

Y1 - 2020/7

N2 - Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and the impact of spoofing and CMs upon ASV and its use as a primary metric in traditional ASV research has long been abandoned in favour of risk-based approaches to assessment. This paper presents several new extensions to the tandem detection cost function (t-DCF), a recent risk-based approach to assess the reliability of spoofing CMs deployed in tandem with an ASV system. Extensions include a simplified version of the t-DCF with fewer parameters, an analysis of a special case for a fixed ASV system, simulations which give original insights into its interpretation and new analyses using the ASVspoof 2019 database. It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.

AB - Recent years have seen growing efforts to develop spoofing countermeasures (CMs) to protect automatic speaker verification (ASV) systems from being deceived by manipulated or artificial inputs. The reliability of spoofing CMs is typically gauged using the equal error rate (EER) metric. The primitive EER fails to reflect application requirements and the impact of spoofing and CMs upon ASV and its use as a primary metric in traditional ASV research has long been abandoned in favour of risk-based approaches to assessment. This paper presents several new extensions to the tandem detection cost function (t-DCF), a recent risk-based approach to assess the reliability of spoofing CMs deployed in tandem with an ASV system. Extensions include a simplified version of the t-DCF with fewer parameters, an analysis of a special case for a fixed ASV system, simulations which give original insights into its interpretation and new analyses using the ASVspoof 2019 database. It is hoped that adoption of the t-DCF for the CM assessment will help to foster closer collaboration between the anti-spoofing and ASV research communities.

KW - Automatic speaker verification (ASV)

KW - detect- ion cost function

KW - presentation attack detection

KW - spoofing counter- measures

UR - http://www.scopus.com/inward/record.url?scp=85089944487&partnerID=8YFLogxK

U2 - 10.1109/TASLP.2020.3009494

DO - 10.1109/TASLP.2020.3009494

M3 - Journal article

AN - SCOPUS:85089944487

SN - 2329-9290

VL - 28

SP - 2195

EP - 2210

JO - IEEE/ACM Transactions on Audio Speech and Language Processing

JF - IEEE/ACM Transactions on Audio Speech and Language Processing

M1 - 9143410

ER -

Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals

Abstract

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this