Enabling On-Device Training of Speech Recognition Models With Federated Dropout

Dhruv Guliani, Lillian Zhou, Chang-Wan Ryu, Tien-Ju Yang, Harry Zhang, Yonghui Xiao, Françoise Beaufays, Giovanni Motta

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) · 2022 · 11 citations · 24 references

Concepts

Abstract

Federated learning can be used to train machine learning models on the edge on local data that never leave devices, providing privacy by default. This presents a challenge pertaining to the communication and computation costs associated with clients’ devices. These costs are strongly correlated with the size of the model being trained, and are significant for state-of-the-art automatic speech recognition models.We propose using federated dropout to reduce the size of client models while training a full-size model server-side. We provide empirical evidence of the effectiveness of federated dropout, and propose a novel approach to vary the dropout rate applied at each layer. Furthermore, we find that federated dropout enables a set of smaller sub-models within the larger model to independently have low word error rates, making it easier to dynamically adjust the size of the model deployed for inference.

References

24