Training Keyword Spotting Models on Non-IID Data with Federated Learning

Andrew Hard, Kurt Partridge, Cameron Nguyen, Niranjan Subrahmanya, Aishanee Shah, Pai Zhu, Ignacio Lopez Moreno, Rajiv Mathews

We demonstrate that a production-quality keyword-spotting model can be trained on-device using federated learning and achieve comparable false accept and false reject rates to a centrally-trained model. To overcome the algorithmic constraints associated with fitting on-device data (which are inherently non-independent and identically distributed), we conduct thorough empirical studies of optimization algorithms and hyperparameter configurations using large-scale federated simulations. To overcome resource constraints, we replace memory-intensive MTR data augmentation with SpecAugment, which reduces the false reject rate by 56%. Finally, to label examples (given the zero visibility into on-device data), we explore teacher-student training.

 DOI: 10.21437/Interspeech.2020-3023

Cite as: Hard, A., Partridge, K., Nguyen, C., Subrahmanya, N., Shah, A., Zhu, P., Moreno, I.L., Mathews, R. (2020) Training Keyword Spotting Models on Non-IID Data with Federated Learning. Proc. Interspeech 2020, 4343-4347, DOI: 10.21437/Interspeech.2020-3023.

  author={Andrew Hard and Kurt Partridge and Cameron Nguyen and Niranjan Subrahmanya and Aishanee Shah and Pai Zhu and Ignacio Lopez Moreno and Rajiv Mathews},
  title={{Training Keyword Spotting Models on Non-IID Data with Federated Learning}},
  booktitle={Proc. Interspeech 2020},