Anti-spoofing Methods for Automatic SpeakerVerification System

Authors: Galina Lavrentyeva, Sergey Novoselov, Konstantin Simonchik

Published: 2017-05-24 16:58:03+00:00

Comment: 12 pages, 0 figures, published in Springer Communications in Computer and Information Science (CCIS) vol. 661

AI Summary

This paper overviews and experimentally compares various acoustic feature spaces and classifiers for robust anti-spoofing countermeasures against Automatic Speaker Verification (ASV) spoofing attacks. It evaluates several spoofing detection systems on the ASVspoof Challenge 2015 datasets, highlighting effective combinations of features and classifiers. Key findings emphasize the importance of magnitude and phase information, wavelet-based features, and the strong performance of SVM and deep neural network classifiers.

Abstract

Growing interest in automatic speaker verification (ASV)systems has lead to significant quality improvement of spoofing attackson them. Many research works confirm that despite the low equal er-ror rate (EER) ASV systems are still vulnerable to spoofing attacks. Inthis work we overview different acoustic feature spaces and classifiersto determine reliable and robust countermeasures against spoofing at-tacks. We compared several spoofing detection systems, presented so far,on the development and evaluation datasets of the Automatic SpeakerVerification Spoofing and Countermeasures (ASVspoof) Challenge 2015.Experimental results presented in this paper demonstrate that the useof magnitude and phase information combination provides a substantialinput into the efficiency of the spoofing detection systems. Also wavelet-based features show impressive results in terms of equal error rate. Inour overview we compare spoofing performance for systems based on dif-ferent classifiers. Comparison results demonstrate that the linear SVMclassifier outperforms the conventional GMM approach. However, manyresearchers inspired by the great success of deep neural networks (DNN)approaches in the automatic speech recognition, applied DNN in thespoofing detection task and obtained quite low EER for known and un-known type of spoofing attacks.

Key findings

Experimental results demonstrate that combining magnitude and phase information, as well as wavelet-based features (MWPC), provides substantial efficiency improvements in spoofing detection, with MWPC showing particularly low EERs. Linear SVM classifiers generally outperform conventional GMMs, and deep neural networks also achieve very low Equal Error Rates (EERs) for both known and unknown types of spoofing attacks. The most efficient systems often leverage complementary information from multiple feature types.

Approach

The authors conduct an overview and experimental comparison of different spoofing detection systems. They investigate various acoustic feature extraction methods, including magnitude-based (MFCC, MFPC), phase-based (GD, MGD, IF, CosPhasePC), and wavelet-based (MWPC) features, and evaluate their performance with different classifiers such as GMM, SVM, and DNNs on the ASVspoof Challenge 2015 dataset.

Datasets

ASVspoof Challenge 2015 (development and evaluation datasets)

Model(s)

GMM (Gaussian Mixture Model), SVM (Support Vector Machine, specifically linear SVM), DNN (Deep Neural Network), DBN (Deep Belief Network), MLP (Multi-Layer Perceptron)

Author countries

Russia

← Previous