Impact of Channel Variation on One-Class Learning for Spoof Detection

TL;DR

This paper investigates how channel variations, such as codec parameters, affect one-class learning-based spoof detection systems and explores training strategies to improve robustness against such variability.

Contribution

It provides an analysis of the impact of codec-induced channel variations on one-class spoof detection and evaluates training strategies like multi-conditional training and mini-batching for robustness.

Findings

Codec parameters significantly impact system performance.

Multi-conditional training improves robustness by 35.55%.

Custom mini-batching captures more generalized features.

Abstract

Margin-based losses, especially one-class classification loss, have improved the generalization capabilities of countermeasure systems (CMs), but their reliability is not tested with spoofing attacks degraded with channel variation. Our experiments aim to tackle this in two ways: first, by investigating the impact of various codec simulations and their corresponding parameters, namely bit-rate, discontinuous transmission (DTX), and loss, on the performance of the one-class classification-based CM system; second, by testing the efficacy of the various settings of margin-based losses for training and evaluating our CM system on codec simulated data. Multi-conditional training (MCT) along with various data-feeding and custom mini-batching strategies were also explored to handle the added variability in the new data setting and to find an optimal setting to carry out the above experiments. Our experimental results reveal that a strict restrain over the embedding space degrades the performance of the one-class classification model. MCT relatively improves performance by 35.55\%, and custom mini-batching captures more generalized features for the new data setting. Whereas varying the codec parameters made a significant impact on the performance of the countermeasure system.