2022/02/17 by Chao-Han Huck Yang, Yang, Chao-Han Huck, Zeeshan Ahmed +13
Computer Science · #Adversarial Robustness in Machine Learning #Anomaly Detection Techniques and Applications #Artificial Intelligence (cs.AI) #Audio and Speech Processing (eess.AS) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Neural and Evolutionary Computing (cs.NE) #Sound (cs.SD) #Speech Recognition and Synthesis #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2202.08532
openalex publication_date 2022/02/17 · openalex created_date 2022/05/05 · openalex updated_date 2026/07/28
In this work, we aim to enhance the system robustness of end-to-end automatic speech recognition (ASR) against adversarially-noisy speech examples. We focus on a rigorous and empirical "closed-model adversarial robustness" setting (e.g., on-device or cloud applications). The adversarial noise is only generated by closed-model optimization (e.g., evolutionary and zeroth-order estimation) without accessing gradient information of a targeted ASR model directly. We propose an advanced Bayesian neural network (BNN) based adversarial detector, which could model latent distributions against adaptive adversarial perturbation with divergence measurement. We further simulate deployment scenarios of RNN Transducer, Conformer, and wav2vec-2.0 based ASR systems with the proposed adversarial detection system. Leveraging the proposed BNN based detection system, we improve detection rate by +2.77 to +5.42% (relative +3.03 to +6.26%) and reduce the word error rate by 5.02 to 7.47% on LibriSpeech datasets compared to the current model enhancement methods against the adversarial speech examples.