We propose using an adversarial autoencoder (AAE) to replace generative adversarial network (GAN) in private aggregation of teacher ensembles (PATE), a solution for ensuring differential privacy in speech applications. The AAE architecture allows us to obtain good synthetic speech leveraging upon a discriminative training of latent vectors. Such synthetic speech is used to build a privacy-preserving classifier when non-sensitive data is not sufficiently available in the public domain. This classifier follows the PATE scheme that uses an ensemble of noisy outputs to label the synthetic samples and guarantee ε-differential privacy (DP) on its derived classifiers. Our proposed framework thus consists of an AAE-based generator and a PATE-based classifier (PATE-AAE). Evaluated on the Google Speech Commands Dataset Version II, the proposed PATE-AAE improves the average classification accuracy by +2.11% and +6.60%, respectively, when compared with alternative privacy-preserving solutions, namely PATE-GAN and DP-GAN, while maintaining a strong level of privacy target at ε=0.01 with a fixed δ=10-5.
Cite as: Yang, C.-H.H., Siniscalchi, S.M., Lee, C.-H. (2021) PATE-AAE: Incorporating Adversarial Autoencoder into Private Aggregation of Teacher Ensembles for Spoken Command Classification. Proc. Interspeech 2021, 881-885, doi: 10.21437/Interspeech.2021-640
@inproceedings{yang21_interspeech,
author={Chao-Han Huck Yang and Sabato Marco Siniscalchi and Chin-Hui Lee},
title={{PATE-AAE: Incorporating Adversarial Autoencoder into Private Aggregation of Teacher Ensembles for Spoken Command Classification}},
year=2021,
booktitle={Proc. Interspeech 2021},
pages={881--885},
doi={10.21437/Interspeech.2021-640}
}