Training Keyword Spotting Models on Non-IID Data with Federated Learning

05/21/2020

∙

by   Andrew Hard, et al.

∙

0

∙

share

We demonstrate that a production-quality keyword-spotting model can be trained on-device using federated learning and achieve comparable false accept and false reject rates to a centrally-trained model. To overcome the algorithmic constraints associated with fitting on-device data (which are inherently non-independent and identically distributed), we conduct thorough empirical studies of optimization algorithms and hyperparameter configurations using large-scale federated simulations. To overcome resource constraints, we replace memory intensive MTR data augmentation with SpecAugment, which reduces the false reject rate by 56 visibility into on-device data), we explore teacher-student training.