Explainable Speech Emotion Recognition: Weighted Attribute Fairness to Model Demographic Contributions to Social Bias

ArXi:2604.19763v1 Announce Type: cross Speech Emotion Recognition (SER) systems have growing applications in sensitive domains such as mental health and education, where biased predictions can cause harm. Traditional fairness metrics, such as Equalised Odds and graphic Parity, often overlook the joint dependency between graphic attributes and model predictions. We propose a fairness modelling approach for SER that explicitly captures allocative bias by learning the joint relationship between graphic attributes and model error.