Explainable Attribute-Based Speaker Verification

TL;DR

This paper introduces an explainable speaker verification system that uses automatically extracted personal attributes like gender, nationality, and age to identify speakers, aiming for transparency and human-like reasoning.

Contribution

It presents a novel attribute-based approach to speaker verification that enhances interpretability while maintaining competitive performance.

Findings

Comparable performance to ground truth with correct attributes

System sacrifices some accuracy for explainability

Lays groundwork for future attribute expansion

Abstract

This paper proposes a fully explainable approach to speaker verification (SV), a task that fundamentally relies on individual speaker characteristics. The opaque use of speaker attributes in current SV systems raises concerns of trust. Addressing this, we propose an attribute-based explainable SV system that identifies speakers by comparing personal attributes such as gender, nationality, and age extracted automatically from voice recordings. We believe this approach better aligns with human reasoning, making it more understandable than traditional methods. Evaluated on the Voxceleb1 test set, the best performance of our system is comparable with the ground truth established when using all correct attributes, proving its efficacy. Whilst our approach sacrifices some performance compared to non-explainable methods, we believe that it moves us closer to the goal of transparent, interpretable AI and lays the groundwork for future enhancements through attribute expansion.