Training Speech Recognition Models With Federated Learning: A Quality/cost Framework

Abstract

We propose using federated learning, a decentralized on-device learning paradigm, to train speech recognition models. By performing epochs of training on a per-user basis, federated learning must incur the cost of dealing with non-IID data distributions, which are expected to negatively affect the quality of the trained model. We propose a framework by which the degree of non-IID-ness can be varied, consequently illustrating a trade-off between model quality and the computational cost of federated training, which we capture through a novel metric. Finally, we demonstrate that hyper-parameter optimization and appropriate use of variational noise are sufficient to compensate for the quality impact of non-IID distributions, while decreasing the cost.

Stats

• citations52
• S2 citations—
• github stars0
• HF likes0
• heat score12.93
• arxiv keyguliani2020training

Related papers

• Fedspeech: Federated Text-to-speech With Continual Learning (2021)9.23
• Federated Pruning: Improving Neural Network Efficiency With Federated Learning (2022)7.50
• Communication-efficient Personalized Federated Learning For Speech-to-text Tasks (2024)7.81
• The Gift Of Feedback: Improving ASR Model Quality By Learning From User Corrections Through Federated Learning (2023)0.00
• Federated Learning For Keyword Spotting (2018)17.09
• Separate But Together: Unsupervised Federated Learning For Speech Enhancement From Non-iid Data (2021)8.35
• Fednst: Federated Noisy Student Training For Automatic Speech Recognition (2022)6.77
• Toward Domain-invariant Speech Recognition Via Large Scale Training (2018)13.39