SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks

Zhizheng Wu; Ali Khodabakhsh; Cenk Demiroglu; Junichi Yamagishi; Daisuke Saito; Tomoki Toda; Simon King

doi:10.1109/ICASSP.2015.7178810

SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks

Zhizheng Wu, Ali Khodabakhsh, Cenk Demiroglu, Junichi Yamagishi, Daisuke Saito, Tomoki Toda, Simon King

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract / Description of output

This paper presents the first version of a speaker verification spoofing and anti-spoofing database, named SAS corpus. The corpus includes nine spoofing techniques, two of which are speech synthesis, and seven are voice conversion. We design two protocols, one for standard speaker verification evaluation, and the other for producing spoofing materials. Hence, they allow the speech synthesis community to produce spoofing materials incrementally without knowledge of speaker verification spoofing and anti-spoofing. To provide a
set of preliminary results, we conducted speaker verification experiments using two state-of-the-art systems. Without any anti-spoofing techniques, the two systems are extremely vulnerable to the spoofing attacks implemented in our SAS corpus.

Original language	English
Title of host publication	Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on
Publisher	Institute of Electrical and Electronics Engineers
Pages	4440-4444
Number of pages	5
DOIs	https://doi.org/10.1109/ICASSP.2015.7178810
Publication status	Published - 2015
Event	IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) - Brisbane, Australia, Australia Duration: 19 Apr 2015 → 24 Apr 2015

Conference

Conference	IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Country/Territory	Australia
City	Brisbane, Australia
Period	19/04/15 → 24/04/15

Access to Document

10.1109/ICASSP.2015.7178810

masterAccepted author manuscript, 142 KB

http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=7178810

Deep architectures for statistical speech synthesis
Yamagishi, J.
UK industry, commerce and public corporations
4/09/12 → 3/03/16
Project: Research
Natural Speech Technology
Renals, S. & King, S.
EPSRC
1/05/11 → 31/07/16
Project: Research

Spoofing and Anti-Spoofing (SAS) corpus v1.0
Wu, Z. (Creator), Khodabakhsh, A. (Creator), Yamagishi, J. (Creator), Saito, D. (Creator), Toda, T. (Creator), Ling, Z. (Creator), King, S. (Creator) & Demiroglu, C. (Creator), Edinburgh DataShare, 27 May 2015
DOI: 10.7488/ds/252, http://www.zhizheng.org/papers/icassp2015_sas.pdf
Dataset

Cite this

Wu, Z., Khodabakhsh, A., Demiroglu, C., Yamagishi, J., Saito, D., Toda, T., & King, S. (2015). SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks. In Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on (pp. 4440-4444). Institute of Electrical and Electronics Engineers. https://doi.org/10.1109/ICASSP.2015.7178810

@inproceedings{079e3e14451f4e838e78fec56a18edd0,

title = "SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks",

abstract = "This paper presents the first version of a speaker verification spoofing and anti-spoofing database, named SAS corpus. The corpus includes nine spoofing techniques, two of which are speech synthesis, and seven are voice conversion. We design two protocols, one for standard speaker verification evaluation, and the other for producing spoofing materials. Hence, they allow the speech synthesis community to produce spoofing materials incrementally without knowledge of speaker verification spoofing and anti-spoofing. To provide aset of preliminary results, we conducted speaker verification experiments using two state-of-the-art systems. Without any anti-spoofing techniques, the two systems are extremely vulnerable to the spoofing attacks implemented in our SAS corpus.",

author = "Zhizheng Wu and Ali Khodabakhsh and Cenk Demiroglu and Junichi Yamagishi and Daisuke Saito and Tomoki Toda and Simon King",

year = "2015",

doi = "10.1109/ICASSP.2015.7178810",

language = "English",

pages = "4440--4444",

booktitle = "Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on",

publisher = "Institute of Electrical and Electronics Engineers",

address = "United States",

note = "IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) ; Conference date: 19-04-2015 Through 24-04-2015",

}

Wu, Z, Khodabakhsh, A, Demiroglu, C, Yamagishi, J, Saito, D, Toda, T & King, S 2015, SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks. in Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on . Institute of Electrical and Electronics Engineers, pp. 4440-4444, IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Brisbane, Australia, Australia, 19/04/15. https://doi.org/10.1109/ICASSP.2015.7178810

SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks. / Wu, Zhizheng; Khodabakhsh, Ali; Demiroglu, Cenk et al.
Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on . Institute of Electrical and Electronics Engineers, 2015. p. 4440-4444.

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks

AU - Wu, Zhizheng

AU - Khodabakhsh, Ali

AU - Demiroglu, Cenk

AU - Yamagishi, Junichi

AU - Saito, Daisuke

AU - Toda, Tomoki

AU - King, Simon

PY - 2015

Y1 - 2015

N2 - This paper presents the first version of a speaker verification spoofing and anti-spoofing database, named SAS corpus. The corpus includes nine spoofing techniques, two of which are speech synthesis, and seven are voice conversion. We design two protocols, one for standard speaker verification evaluation, and the other for producing spoofing materials. Hence, they allow the speech synthesis community to produce spoofing materials incrementally without knowledge of speaker verification spoofing and anti-spoofing. To provide aset of preliminary results, we conducted speaker verification experiments using two state-of-the-art systems. Without any anti-spoofing techniques, the two systems are extremely vulnerable to the spoofing attacks implemented in our SAS corpus.

AB - This paper presents the first version of a speaker verification spoofing and anti-spoofing database, named SAS corpus. The corpus includes nine spoofing techniques, two of which are speech synthesis, and seven are voice conversion. We design two protocols, one for standard speaker verification evaluation, and the other for producing spoofing materials. Hence, they allow the speech synthesis community to produce spoofing materials incrementally without knowledge of speaker verification spoofing and anti-spoofing. To provide aset of preliminary results, we conducted speaker verification experiments using two state-of-the-art systems. Without any anti-spoofing techniques, the two systems are extremely vulnerable to the spoofing attacks implemented in our SAS corpus.

U2 - 10.1109/ICASSP.2015.7178810

DO - 10.1109/ICASSP.2015.7178810

M3 - Conference contribution

SP - 4440

EP - 4444

BT - Acoustics, Speech and Signal Processing (ICASSP), 2015 IEEE International Conference on

PB - Institute of Electrical and Electronics Engineers

T2 - IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Y2 - 19 April 2015 through 24 April 2015

ER -

SAS: A Speaker Verification Spoofing Database Containing Diverse Attacks

Abstract / Description of output

Conference

Access to Document

Fingerprint

Projects

Deep architectures for statistical speech synthesis

Natural Speech Technology

Datasets

Spoofing and Anti-Spoofing (SAS) corpus v1.0 ﻿

Cite this

Spoofing and Anti-Spoofing (SAS) corpus v1.0