Real-Time, Universal, and Robust Adversarial Attacks Against Speaker Recognition Systems
Authors:
- 1 May 2020
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1738-1742
Abstract
As the popularity of voice user interface (VUI) exploded in recent years, speaker recognition system has emerged as an important medium of identifying a speaker in many security-required applications and services. In this paper, we propose the first real-time, universal, and robust adversarial attack against the state-of-the-art deep neural network (DNN) based speaker recognition system. Through adding an audio-agnostic universal perturbation on arbitrary enrolled speaker’s voice input, the DNN-based speaker recognition system would identify the speaker as any target (i.e., adversary-desired) speaker label. In addition, we improve the robustness of our attack by modeling the sound distortions caused by the physical over-the-air propagation through estimating room impulse response (RIR). Experiment using a public dataset of 109 English speakers demonstrates the effectiveness and robustness of our proposed attack with a high attack success rate of over 90%. The attack launching time also achieves a 100× speedup over contemporary non-universal attacks.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- Probing the Information Encoded in X-Vectors2019
- Universal Adversarial Perturbations for Speech Recognition Systems2019
- Speaker Recognition for Multi-speaker Conversations Using X-vectors2019
- Audio Adversarial Examples: Targeted Attacks on Speech-to-Text2018
- Fooling End-To-End Speaker Verification With Adversarial Examples2018
- Pyroomacoustics: A Python Package for Audio Room Simulation and Array Processing Algorithms2018
- X-Vectors: Robust DNN Embeddings for Speaker Recognition2018
- Deep Neural Network Embeddings for Text-Independent Speaker Verification2017
- Advances in deep neural network approaches to speaker recognition2015
- A novel scheme for speaker recognition using a phonetically-aware deep neural network2014