Adversarial Audio

Adversarial audio research focuses on creating subtly altered audio files—imperceptible to humans but designed to fool automatic speech recognition (ASR) and speaker verification (SV) systems. Current research explores various attack methods, including those leveraging neural networks (e.g., Hammerstein models) to generate perturbations, linguistic features to manipulate transcriptions efficiently, and techniques to create universal adversarial segments affecting multiple models. This field is crucial for understanding and mitigating security vulnerabilities in voice-controlled devices and biometric systems, impacting the development of robust and trustworthy AI applications.

Papers

August 17, 2024

Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model
Massimiliano Todisco, Michele Panariello, Xin Wang, Héctor Delgado, Kong Aik Lee, Nicholas Evans
Adversarial Attack Adversarial Perturbation Speaker Verification Speaker Embeddings Adversarial Audio

August 3, 2024

ALIF: Low-Cost Adversarial Audio Attacks on Black-Box Speech Platforms using Linguistic Features
Peng Cheng, Yuwei Wang, Peng Huang, Zhongjie Ba, Xiaodong Lin, Feng Lin, Li Lu, Kui Ren
Adversarial Attack Adversarial Example Linguistic Feature Black Box Language Model Adversarial Audio

July 5, 2024

Controlling Whisper: Universal Acoustic Adversarial Attacks to Control Speech Foundation Models
Vyas Raina, Mark Gales
Large Language Model State of the Art Whisper Speech Foundation Model Adversarial Audio

May 9, 2024

Muting Whisper: A Universal Acoustic Adversarial Attack on Speech Foundation Models
Vyas Raina, Rao Ma, Charles McGhee, Kate Knill, Mark Gales
Adversarial Attack State of the Art Whisper Speech Foundation Model Adversarial Audio Open Whisper Style Speech Model Speech Privacy

October 5, 2023

An Integrated Algorithm for Robust and Imperceptible Audio Adversarial Examples
Armin Ettenhofer, Jan-Philipp Schulze, Karla Pizzi
Automatic Speech Recognition Adversarial Example Robust Version Impulse Response Adversarial Audio Psychoacoustic Model Unified Algorithm Audio Adversarial Example

September 13, 2023

PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection
Hanqing Guo, Guangjing Wang, Yuanda Wang, Bocheng Chen, Qiben Yan, Li Xiao
Adversarial Attack Black Box Black Box Attack Sound Source Adversarial Audio

September 2, 2023

Timbre-reserved Adversarial Attack in Speaker Identification
Qing Wang, Jixun Yao, Li Zhang, Pengcheng Guo, Lei Xie
Adversarial Attack Speaker Identification Adversarial Audio

August 18, 2023

Compensating Removed Frequency Components: Thwarting Voice Spectrum Reduction Attacks
Shu Wang, Kun Sun, Qi Li
Adversarial Audio Frequency Estimation Spectral Domain Attack Voice Cloning Attack Audio Attack

May 26, 2023

DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution
Matías P. Pizarro B., Dorothea Kolossa, Asja Fischer
Adversarial Attack Automatic Speech Recognition Adversarial Example Automatic Speech Recognition System Distinct AInality Trait Adversarial Data Adversarial Audio Output Distribution

April 20, 2023

Towards the Universal Defense for Query-Based Audio Adversarial Attacks
Feng Guo, Zheng Sun, Yuxuan Chen, Lei Ju
Adversarial Audio Audio Adversarial Example

April 18, 2023

Towards the Transferable Audio Adversarial Attack via Ensemble Methods
Feng Guo, Zheng Sun, Yuxuan Chen, Lei Ju
Deep Learning Adversarial Attack Deep Learning Model Adversarial Example Ensemble Method Transferable Attack Adversarial Audio

March 28, 2023

TransAudio: Towards the Transferable Adversarial Audio Attack via Learning Contextualized Perturbations
Qi Gege, Yuefeng Chen, Xiaofeng Mao, Yao Zhu, Binyuan Hui, Xiaodan Li, Rong Zhang, Hui Xue
Adversarial Example Transferable Attack Adversarial Audio Word Level Attack Transfer Based Attack Context Perturbation

March 2, 2023

Defending against Adversarial Audio via Diffusion Model
Shutong Wu, Jiongxiao Wang, Wei Ping, Weili Nie, Chaowei Xiao
Diffusion Model Adversarial Attack Adversarial Training Adversarial Purification Adversarial Audio Audio Adversarial Example

November 17, 2022

Privacy against Real-Time Speech Emotion Detection via Acoustic Adversarial Evasion of Machine Learning
Brian Testa, Yi Xiao, Harshit Sharma, Avery Gump, Asif Salekin
Machine Learning Speech Emotion Recognition Privacy Policy Voice Assistant Speech Transcription Imperceptible Adversarial Perturbation Adversarial Audio Smart Speaker

October 21, 2022

Adaptive re-calibration of channel-wise features for Adversarial Audio Classification
Vardhan Dongre, Abhinav Thimma Reddy, Nikhitha Reddeddy
Synthesized Speech Deepfake Audio Channel Wise Synthetic Speech Detection Adversarial Audio Synthetic Speech Detector Adaptive Calibration

September 21, 2022

Mandarin Singing Voice Synthesis with Denoising Diffusion Probabilistic Wasserstein GAN
Yin-Ping Cho, Yu Tsao, Hsin-Min Wang, Yi-Wen Liu
Denoising Diffusion Neural Vocoder Singing Voice Singing Voice Synthesis Synthetic Voice Adversarial Audio

July 26, 2022

Perception-Aware Attack: Creating Adversarial Music via Reverse-Engineering Human Perception
Rui Duan, Zhe Qu, Shangqing Zhao, Leah Ding, Yao Liu, Zhuo Lu
Adversarial Attack Human Perception Adversarial Audio Man in the Middle Attack

June 7, 2022

AS2T: Arbitrary Source-To-Target Adversarial Attack on Speaker Recognition Systems
Guangke Chen, Zhe Zhao, Fu Song, Sen Chen, Lingling Fan, Yang Liu
Adversarial Attack Speaker Recognition System Adversarial Audio

March 18, 2022

Neural Predictor for Black-Box Adversarial Attacks on Speech Recognition
Marie Biolková, Bac Nguyen
Adversarial Attack Adversarial Example Speech Recognition Black Box Adversarial Attack Adversarial Audio Neural Predictor

February 22, 2022

Sound Adversarial Audio-Visual Navigation
Yinfeng Yu, Wenbing Huang, Fuchun Sun, Changan Chen, Yikai Wang, Xiaohong Liu
Adversarial Audio Audio Visual Navigation Audio Attack

Adversarial Audio

Papers

Malacopula: adversarial automatic speaker verification attacks using a neural-based generalised Hammerstein model

ALIF: Low-Cost Adversarial Audio Attacks on Black-Box Speech Platforms using Linguistic Features

Controlling Whisper: Universal Acoustic Adversarial Attacks to Control Speech Foundation Models

Muting Whisper: A Universal Acoustic Adversarial Attack on Speech Foundation Models

An Integrated Algorithm for Robust and Imperceptible Audio Adversarial Examples

PhantomSound: Black-Box, Query-Efficient Audio Adversarial Attack via Split-Second Phoneme Injection

Timbre-reserved Adversarial Attack in Speaker Identification

Compensating Removed Frequency Components: Thwarting Voice Spectrum Reduction Attacks

DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution

Towards the Universal Defense for Query-Based Audio Adversarial Attacks

Towards the Transferable Audio Adversarial Attack via Ensemble Methods

TransAudio: Towards the Transferable Adversarial Audio Attack via Learning Contextualized Perturbations

Defending against Adversarial Audio via Diffusion Model

Privacy against Real-Time Speech Emotion Detection via Acoustic Adversarial Evasion of Machine Learning

Adaptive re-calibration of channel-wise features for Adversarial Audio Classification

Mandarin Singing Voice Synthesis with Denoising Diffusion Probabilistic Wasserstein GAN

Perception-Aware Attack: Creating Adversarial Music via Reverse-Engineering Human Perception

AS2T: Arbitrary Source-To-Target Adversarial Attack on Speaker Recognition Systems

Neural Predictor for Black-Box Adversarial Attacks on Speech Recognition

Sound Adversarial Audio-Visual Navigation