Adversarial Data Augmentation for Robust Speaker Verification
Zhenyu Zhou, Junhui Chen, Namin Wang, Lantian Li, Dong Wang · 2023
Data augmentation (DA) has gained widespread popularity in deep speaker models due to its ease of implementation and significant effectiveness. It enriches training data by simulating real-life acoustic variations, enabling deep neural networks to learn speaker-related representations while disregarding irrelevant acoustic variations, thereby improving robustness and generalization. However, a potential issue with the vanilla DA is augmentation residual, i.e., unwanted distortion caused by different types of augmentation.