arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2310.03349cs.SDcs.LGeess.AS
收藏

一种用于生成鲁棒且不可感知的音频对抗样本的集成算法

An Integrated Algorithm for Robust and Imperceptible Audio Adversarial Examples

Fraunhofer Institute for Applied and Integrated Security (AISEC)(弗劳恩霍夫应用与集成安全研究所(AISEC));Technical University Munich(慕尼黑工业大学)

Armin Ettenhofer, Jan-Philipp Schulze, Karla Pizzi

AI总结:

本研究提出一种集成算法,在音频对抗样本生成阶段就结合心理声学模型与神经网络动态生成的房间冲激响应,提升了样本的不可感知性与空中攻击鲁棒性,同时伴随词错误率上升。

摘要:

音频对抗样本是指经过篡改、能够欺骗自动语音识别(ASR)系统,但对人类听众而言听起来仍属正常的音频文件。大多数生成此类样本的方法都基于两步算法:首先生成可用的对抗音频文件,随后针对可感知性和鲁棒性对其进行微调。在本研究中,我们提出了一种集成算法,该算法在生成阶段就引入了心理声学模型和房间冲激响应(room impulse responses, RIR)。RIR由神经网络在生成过程中动态生成,用于模拟物理环境,从而增强样本的抗变换能力,以应对空中传播攻击中会经历的信号变换。我们通过三项实验对不同方法进行了对比:在模拟环境和真实空中传播场景中评估鲁棒性,通过人类受试者研究评估可感知性。我们仅考虑心理声学、或同时结合心理声学与鲁棒性优化的算法,在信噪比(SNR)以及人类感知研究中均表现出提升,代价是词错误率(WER)有所上升。

英文摘要:

Audio adversarial examples are audio files that have been manipulated to fool an automatic speech recognition (ASR) system, while still sounding benign to a human listener. Most methods to generate such samples are based on a two-step algorithm: first, a viable adversarial audio file is produced, then, this is fine-tuned with respect to perceptibility and robustness. In this work, we present an integrated algorithm that uses psychoacoustic models and room impulse responses (RIR) in the generation step. The RIRs are dynamically created by a neural network during the generation process to simulate a physical environment to harden our examples against transformations experienced in over-the-air attacks. We compare the different approaches in three experiments: in a simulated environment and in a realistic over-the-air scenario to evaluate the robustness, and in a human study to evaluate the perceptibility. Our algorithms considering psychoacoustics only or in addition to the robustness show an improvement in the signal-to-noise ratio (SNR) as well as in the human perception study, at the cost of an increased word error rate (WER).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑