Next Article in Journal
CLASSY: A Conversational Aware Suggestion System
Previous Article in Journal
Smart Environments Design on Industrial Automated Greenhouses
 
 
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Proceeding Paper

On the Use of Fisher Vector Encoding for Voice Spoofing Detection †

Computer Research Institute of Montreal (CRIM), Montréal, QC H3N 1M3, Canada
Presented at the 13th International Conference on Ubiquitous Computing and Ambient Intelligence UCAmI 2019, Toledo, Spain, 2–5 December 2019.‬‬‬‬‬‬‬‬
Proceedings 2019, 31(1), 37; https://doi.org/10.3390/proceedings2019031037
Published: 20 November 2019

Abstract

Recently, the vulnerability of automatic speaker recognition systems to spoofing attacks has received significant interest among researchers. A robust speaker recognition system demands not only high recognition accuracy but also robustness to spoofing attacks. Several spoofing and countermeasure challenges have been organized to draw attention to this problem among the speaker recognition communities. Low-level descriptors designed to detect artifacts in spoofed speech are found to be the most effective countermeasures against spoofing attacks. In this work, we used Fisher vector encoding of low-level descriptors extracted from speech signals. The idea behind Fisher vector encoding is to determine the amount of change induced by the descriptors of the signal on a background probability model which is typically a Gaussian mixture model. The Fisher vector encodes the amount of change of the model parameters to optimally fit the new- coming data. For performance evaluation of the proposed approach we carried out spoofing detection experiments on the 2015 edition of automatic speaker verification spoofing and countermeasure challenge (ASVspoof2015) and report results on the evaluation set. As baseline systems, we used the standard Gaussian mixture model and i-vector/PLDA paradigms. For a fair comparison, in all systems, Constant Q cepstral coefficient (CQCC) features were used as low-level descriptors. With the Fisher vector-based approach, we achieved an equal error rate (EER) of 0.1145% on the known attacks, 1.223% on the unknown attacks, and 0.668% on the average. Moreover, with a single decision threshold this approach yielded an EER of 1.05% on the evaluation set.
Keywords: spoofing detection; CQCC; Fisher vector; GMM; i-vector; PLDA spoofing detection; CQCC; Fisher vector; GMM; i-vector; PLDA

Share and Cite

MDPI and ACS Style

Alam, J. On the Use of Fisher Vector Encoding for Voice Spoofing Detection. Proceedings 2019, 31, 37. https://doi.org/10.3390/proceedings2019031037

AMA Style

Alam J. On the Use of Fisher Vector Encoding for Voice Spoofing Detection. Proceedings. 2019; 31(1):37. https://doi.org/10.3390/proceedings2019031037

Chicago/Turabian Style

Alam, Jahangir. 2019. "On the Use of Fisher Vector Encoding for Voice Spoofing Detection" Proceedings 31, no. 1: 37. https://doi.org/10.3390/proceedings2019031037

Article Metrics

Back to TopTop