How Do Neural Spoofing Countermeasures Detect Partially Spoofed Audio?

TL;DR

This paper investigates how countermeasures detect partially spoofed audio, revealing their focus on transition artifacts and providing interpretability insights to improve detection models.

Contribution

It introduces a quantitative analysis method to interpret countermeasure decisions and explains their focus on transition artifacts in partially spoofed audio.

Findings

Countermeasures focus on transition artifacts at concatenation regions.

Different focus patterns are observed for fully vs. partially spoofed audio.

Insights aid in designing better spoof detection models and datasets.

Abstract

Partially manipulating a sentence can greatly change its meaning. Recent work shows that countermeasures (CMs) trained on partially spoofed audio can effectively detect such spoofing. However, the current understanding of the decision-making process of CMs is limited. We utilize Grad-CAM and introduce a quantitative analysis metric to interpret CMs' decisions. We find that CMs prioritize the artifacts of transition regions created when concatenating bona fide and spoofed audio. This focus differs from that of CMs trained on fully spoofed audio, which concentrate on the pattern differences between bona fide and spoofed parts. Our further investigation explains the varying nature of CMs' focus while making correct or incorrect predictions. These insights provide a basis for the design of CM models and the creation of datasets. Moreover, this work lays a foundation of interpretability in the field of partial spoofed audio detection that has not been well explored previously.