Adversarial Attacks on Audio Source Separation

Inoue, Shota; Mitsufuji, Yuki

doi:10.1109/icassp39728.2021.9414844

Cited by 6 publications

(2 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The aim is to develop a method N that creates a noise 𝑛 to be applied to the input 𝑥 in order to transform 𝑥 into 𝑥 ′ so that it satisfies optimal performance in system S and degraded performance in system A. [22]. Expanding on previous work in the area of audio scrambling [13], we suggest increased research to explore adversarial AI components that check whether speech content and individual identity could be reconstructed from degraded or scrambled audio.…”

Section: Research Directions and Challenges 31 Reliability: Data Degr...mentioning

confidence: 99%

Safe audio AI services in smart buildings

Williams

Yazdanpanah

Stein

2022

Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation

View full text Add to dashboard Cite

Audio AI services present an opportunity to conceptualise smart buildings in a new light. Microphones can capture fine-grained audio information that can be used for determining how many people are inside of a building, where they are, and what kinds of activities are taking place. This information can feed into smart resource management systems or it could be used for assistive technologies. Generally speaking, audio is regarded as a less intrusive type of information collection than video surveillance, but significant issues of privacy and security persist with audio capture. Such issues warrant a serious discussion about how safe it is to use audio-capture in smart buildings for AI decision-making. This position paper initiates a discussion of research directions for the safety of audio services related to three key areas: data degradation strategies, dynamic customisation of tools, and privacy-aware technologies. In each area, we identify key challenges and highlight solution concepts with the potential to address the issue. CCS CONCEPTS• Security and privacy → Social aspects of security and privacy; • Computing methodologies → Adversarial learning; Speech recognition; • Computer systems organization → Embedded software.

show abstract

Section: Research Directions and Challenges 31 Reliability: Data Degr...mentioning

confidence: 99%

Safe audio AI services in smart buildings

Williams

Yazdanpanah

Stein

2022

Proceedings of the 9th ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation

View full text Add to dashboard Cite

show abstract

“…Kos et al [7] proposed an adversarial example to attack the image reconstruction model, which is the first work related to the deep generation model. In the audio domain, Takahashi et al proposed an adversarial example to attack an audio source separation model that can help protect songs' copyrights from the abuse of separated signals [8]. Huang et al proposed work of attacks on voice conversion systems [9], which can protect a speaker's private information.…”

Section: Introductionmentioning

confidence: 99%

Adversarial Examples Protect Your Privacy on Speech Enhancement System

Dong¹,

Yan²,

Wang³

2023

Computer Systems Science and Engineering

View full text Add to dashboard Cite

Speech is easily leaked imperceptibly. When people use their phones, the personal voice assistant is constantly listening and waiting to be activated. Private content in speech may be maliciously extracted through automatic speech recognition (ASR) technology by some applications on phone devices. To guarantee that the recognized speech content is accurate, speech enhancement technology is used to denoise the input speech. Speech enhancement technology has developed rapidly along with deep neural networks (DNNs), but adversarial examples can cause DNNs to fail. Considering that the vulnerability of DNN can be used to protect the privacy in speech. In this work, we propose an adversarial method to degrade speech enhancement systems, which can prevent the malicious extraction of private information in speech. Experimental results show that the generated enhanced adversarial examples can be removed most content of the target speech or replaced with target speech content by speech enhancement. The word error rate (WER) between the enhanced original example and enhanced adversarial example recognition result can reach 89.0%. WER of target attack between enhanced adversarial example and target example is low at 33.75%. The adversarial perturbation in the adversarial example can bring much more change than itself. The rate of difference between two enhanced examples and adversarial perturbation can reach more than 1.4430. Meanwhile, the transferability between different speech enhancement models is also investigated. The low transferability of the method can be used to ensure the content in the adversarial example is not damaged, the useful information can be extracted by the friendly ASR. This work can prevent the malicious extraction of speech.

show abstract

Amicable Examples for Informed Source Separation

Mitsufuji¹

2022

ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

Adversarial Attacks on Audio Source Separation

Cited by 6 publications

References 21 publications

Safe audio AI services in smart buildings

Safe audio AI services in smart buildings

Adversarial Examples Protect Your Privacy on Speech Enhancement System

Amicable Examples for Informed Source Separation

Contact Info

Product

Resources

About