Towards Evaluating the Robustness of Automatic Speech Recognition Systems via Audio Style Transfer
Not on the map. Why?
Abstract (arXiv)
In light of the widespread application of Automatic Speech Recognition (ASR) systems, their security concerns have received much more attention than ever before, primarily due to the susceptibility of Deep Neural Networks. Previous studies have illustrated that surreptitiously crafting adversarial perturbations enables the manipulation of speech recognition systems, resulting in the production of malicious commands. These attack methods mostly require adding noise perturbations under ℓp norm constraints, inevitably leaving behind artifacts of manual modifications. Recent research has alleviated this limitation by manipulating style vectors to synthesize adversarial examples based on Text-to-Speech (TTS) synthesis audio. However, style modifications based on optimization objectives significantly reduce the controllability and editability of audio styles. In this paper, we propose an attack on ASR systems based on user-customized style transfer. We first test the effect of Style Transfer Attack (STA) which combines style transfer and adversarial attack in sequential order. And then, as an improvement, we propose an iterative Style Code Attack (SCA) to maintain audio quality. Experimental results show that our method can meet the need for user-customized styles and achieve a success rate of 82% in attacks, while keeping sound naturalness due to our user study.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 24 of its 47 references that are in the map. First posted on arXiv 15 May 2024.
Builds on
24 of 47 references.
References · 24
- 2016Towards Evaluating the Robustness of Neural Networkscited by 2,976
- 2015Deep Speech 2: End-to-End Speech Recognition in English and Mandarincited by 449
- 2019AUTOVC: Zero-Shot Voice Style Transfer with Only Autoencoder Losscited by 189
- 2019Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognitioncited by 123
- 2020Unsupervised Speech Decomposition via Triple Information Bottleneckcited by 87
- 2018Adversarial Attacks Against Automatic Speech Recognition Systems via Psychoacoustic Hidingcited by 83
- 2023A Review of Deep Learning Techniques for Speech Processingcited by 52
- 2020SoK: The Faults in our ASRs: An Overview of Attacks against Automatic Speech Recognition and Speaker Identification Systemscited by 46
- 2018Robust Audio Adversarial Example for a Physical Attackcited by 42
- 2018Targeted Adversarial Examples for Black Box Audio Systemscited by 41
- 2021WaveGuard: Understanding and Mitigating Audio Adversarial Examplescited by 39
- 2018Characterizing Audio Adversarial Examples Using Temporal Dependencycited by 34
12 more
- 2019Practical Hidden Voice Attacks against Speech and Speaker Recognition Systemscited by 34
- 2022StyleTTS: A Style-Based Generative Model for Natural and Diverse Text-to-Speech Synthesiscited by 32
- 2020Enabling Fast and Universal Audio Adversarial Attack Using Generative Modelcited by 31
- 2022SpeechSplit 2.0: Unsupervised speech disentanglement for voice conversion Without tuning autoencoder Bottleneckscited by 27
- 2021Black-box Adversarial Attacks on Commercial Speech Platforms with Minimal Informationcited by 25
- 2021STYLER: Style Factor Modeling with Rapidity and Robustness via Speech Decomposition for Expressive and Controllable Neural Text to Speechcited by 18
- 2022StyleFool: Fooling Video Classification Systems via Style Transfercited by 18
- 2017Audio style transfercited by 15
- 2021Fine-grained style control in Transformer-based Text-to-speech Synthesiscited by 15
- 2019Weighted-Sampling Audio Adversarial Example Attackcited by 3
- 2021Towards Robust Speech-to-Text Adversarial Attackcited by 3
- 2022Adversarial Attacks on ASR Systems: An Overviewcited by 1
Built on by
Citing papers per year: 2025: 2.
2 papers.