Jan Silovsky

Enabling Differentially Private Federated Learning for Speech Recognition: Benchmarks, Adaptive Optimizers, and Gradient Clipping

Federated Learning with Differential Privacy for End-to-End Speech Recognition