Debiased Automatic Speech Recognition for Dysarthric Speech via Sample Reweighting with Sample Affinity Test
This addresses fairness and robustness in ASR for dysarthric speakers, an important but often underserved group, though the approach appears incremental as it builds on existing debiasing techniques.
The paper tackled the problem of biased automatic speech recognition (ASR) systems that perform poorly for dysarthric speakers compared to healthy speakers, and introduced a sample reweighting method (Re-SAT) that improved ASR performance on dysarthric speech without degrading it on healthy speech.
Automatic speech recognition systems based on deep learning are mainly trained under empirical risk minimization (ERM). Since ERM utilizes the averaged performance on the data samples regardless of a group such as healthy or dysarthric speakers, ASR systems are unaware of the performance disparities across the groups. This results in biased ASR systems whose performance differences among groups are severe. In this study, we aim to improve the ASR system in terms of group robustness for dysarthric speakers. To achieve our goal, we present a novel approach, sample reweighting with sample affinity test (Re-SAT). Re-SAT systematically measures the debiasing helpfulness of the given data sample and then mitigates the bias by debiasing helpfulness-based sample reweighting. Experimental results demonstrate that Re-SAT contributes to improved ASR performance on dysarthric speech without performance degradation on healthy speech.