Training Speech Recognition Models with Federated Learning: A Quality/Cost Framework
arXiv:2010.15965 · doi:10.1109/ICASSP39728.2021.9413397
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.
Paper published at ICASSP 2021
References in corpus (3)
Cited by in corpus (12)
- Federated Learning Meets Natural Language Processing: A Survey
- Auto-weighted Robust Federated Learning with Corrupted Data Sources
- Open Challenges in Synthetic Speech Detection
- Privacy attacks for automatic speech recognition acoustic models in a federated learning framework
- Separate but Together: Unsupervised Federated Learning for Speech Enhancement from Non-IID Data
- Personalized Federated Learning with Exact Stochastic Gradient Descent
- Federated Learning in ASR: Not as Easy as You Think
- ILASR: Privacy-Preserving Incremental Learning for Automatic Speech Recognition at Production Scale
- Ed-Fed: A generic federated learning framework with resource-aware client selection for edge devices
- Private Language Model Adaptation for Speech Recognition
- RecUP-FL: Reconciling Utility and Privacy in Federated Learning via User-configurable Privacy Defense
- Incremental Layer-wise Self-Supervised Learning for Efficient Speech Domain Adaptation On Device