Spoofing Attack Detection using the Non-linear Fusion of Sub-band Classifiers

View on arXiv ← Back to list

Authors: Hemlata Tak, Jose Patino, Andreas Nautsch, Nicholas Evans, Massimiliano Todisco

Published: 2020-05-20 23:37:28+00:00

AI Summary

This paper proposes a simple yet effective approach for spoofing attack detection in automatic speaker verification. It uses an ensemble of simple classifiers, each tuned to different sub-bands of the audio spectrum, and combines their scores using non-linear fusion, outperforming most systems in the ASVspoof 2019 challenge.

Abstract

The threat of spoofing can pose a risk to the reliability of automatic speaker verification. Results from the bi-annual ASVspoof evaluations show that effective countermeasures demand front-ends designed specifically for the detection of spoofing artefacts. Given the diversity in spoofing attacks, ensemble methods are particularly effective. The work in this paper shows that a bank of very simple classifiers, each with a front-end tuned to the detection of different spoofing attacks and combined at the score level through non-linear fusion, can deliver superior performance than more sophisticated ensemble solutions that rely upon complex neural network architectures. Our comparatively simple approach outperforms all but 2 of the 48 systems submitted to the logical access condition of the most recent ASVspoof 2019 challenge.

Key findings

Non-linear fusion of sub-band classifiers significantly outperforms linear fusion. The proposed simple approach, using GMMs and non-linear fusion, achieves superior performance compared to most of the 48 complex systems submitted to the ASVspoof 2019 challenge. The use of high-resolution front-ends is crucial for effective spoofing detection.

Approach

The approach uses a bank of simple classifiers, each with a front-end optimized for detecting spoofing artifacts in specific sub-bands of the audio spectrum. These classifiers' scores are then non-linearly fused using a Gaussian Mixture Model or Support Vector Machine to achieve a final spoofing detection score.

Datasets

ASVspoof 2019 Logical Access (LA) database

Model(s)

Gaussian Mixture Model (GMM) for individual sub-band classifiers and for fusion; Support Vector Machine (SVM) for fusion.

Author countries

France

← Previous