A Study on Bias and Fairness In Deep Speaker Recognition
This work addresses fairness issues in speaker recognition systems used for authentication and personalization, which is important for ensuring equitable access to smart devices, but it is incremental as it applies existing fairness metrics to this domain.
The study investigated fairness in deep speaker recognition systems by evaluating five neural architectures and five loss functions against gender and nationality groups using three fairness definitions, finding that more sophisticated encoders improve fairness and loss function choice significantly affects bias.
With the ubiquity of smart devices that use speaker recognition (SR) systems as a means of authenticating individuals and personalizing their services, fairness of SR systems has becomes an important point of focus. In this paper we study the notion of fairness in recent SR systems based on 3 popular and relevant definitions, namely Statistical Parity, Equalized Odds, and Equal Opportunity. We examine 5 popular neural architectures and 5 commonly used loss functions in training SR systems, while evaluating their fairness against gender and nationality groups. Our detailed experiments shed light on this concept and demonstrate that more sophisticated encoder architectures better align with the definitions of fairness. Additionally, we find that the choice of loss functions can significantly impact the bias of SR models.