Federating Dynamic Models using Early-Exit Architectures for Automatic Speech Recognition on Heterogeneous Clients

TL;DR

This paper introduces a federated learning approach for speech recognition that uses early-exit dynamic models to adapt to heterogeneous client devices, improving efficiency and privacy.

Contribution

It proposes a novel federated learning method employing early-exit architectures for adaptive, resource-efficient speech recognition on diverse edge devices.

Findings

Effective performance on public datasets

Compatible with basic federated learning strategies

Reduces computational burden on edge devices

Abstract

Automatic speech recognition models require large amounts of speech recordings for training. However, the collection of such data often is cumbersome and leads to privacy concerns. Federated learning has been widely used as an effective decentralized technique that collaboratively learns a shared prediction model while keeping the data local on different clients. Unfortunately, client devices often feature limited computation and communication resources leading to practical difficulties for large models. In addition, the heterogeneity that characterizes edge devices makes it sub-optimal to generate a single model that fits all of them. Differently from the recent literature, where multiple models with different architectures are used, in this work, we propose using dynamical architectures which, employing early-exit solutions, can adapt their processing (i.e. traversed layers) depending on the input and on the operation conditions. This solution falls in the realm of partial training methods and brings two benefits: a single model is used on a variety of devices; federating the models after local training is straightforward. Experiments on public datasets show that our proposed approach is effective and can be combined with basic federated learning strategies.