Robust speaker verification using Population-based Data Augmentation

Speaker recognition under environments with low signal-to-noise ratio (SNR) and high reverberation level has always been challenging. Data augmentation can be applied to simulate the adverse environments that a speaker recognition system may encounter. Typically, the augmentation parameters are manually set. Recently, automatic hyper-parameter optimization using population-based learning has shown promising results. In this paper, we propose a population-based searching strategy for optimizing the augmentation parameters. We refer to the resulting augmentation as population-based augmentation (PBA). Instead of finding a fixed set of hyper-parameters, PBA learns a scheduler for setting the hyper-parameters. This strategy offers a huge computation advantage over the grid search. We were able to obtain high-performance augmentation policies using a population of six networks only. With PBA, we achieved an EER of 3.98% on the VOiCES19 evaluation set.