Adversarial Speaker Distillation for Countermeasure Model on Automatic Speaker Verification

Abstract

The countermeasure (CM) model is developed to protect ASV systems from spoof attacks and prevent resulting personal information leakage in Automatic Speaker Verification (ASV) system. Based on practicality and security considerations, the CM model is usually deployed on edge devices, which have more limited computing resources and storage space than cloud-based systems, confining the model size under a limitation. To better trade off the CM model sizes and performance, we proposed an adversarial speaker distillation method, which is an improved version of knowledge distillation method combined with generalized end-to-end (GE2E) pre-training and adversarial fine-tuning. In the evaluation phase of the ASVspoof 2021 Logical Access task, our proposed adversarial speaker distillation ResNetSE (ASD-ResNetSE) model reaches 0.2695 min t-DCF and 3.54% EER. ASD-ResNetSE only used 22.5% of parameters and 19.4% of multiply and accumulate operands of ResNetSE model.

Accepted by ISCA SPSC 2022

References in corpus (6)

• Distilling the Knowledge in a Neural Network
• In defence of metric learning for speaker recognition
• Tandem Assessment of Spoofing Countermeasures and Automatic Speaker Verification: Fundamentals
• Clova Baseline System for the VoxCeleb Speaker Recognition Challenge 2020
• UR Channel-Robust Synthetic Speech Detection System for ASVspoof 2021
• Improving the Adversarial Robustness for Speaker Verification by Self-Supervised Learning