Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognition

• URL: http://arxiv.org/abs/2606.05678v1
• Date: Thu, 04 Jun 2026 04:00:48 GMT
• Title: Beyond Waveform Robustness: Robust Feature-Vocoder Adversarial Attacks on Automatic Speech Recognition
• Abstract summary: Existing adversarial attacks directly add adversarial noise to the speech audio.<n>Clean-Referenced Feature-Vocoder Attack moves the adversarial search space from raw waveforms to self-supervised learning representations.<n>Experiments show that, when optimized only on raw Whisper-small as a public surrogate model, our attack transfers effectively to black-box ASR models.
• Score: 9.839293944956006
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Automatic speech recognition (ASR) systems have become widely used for multilingual speech-to-text transcription. Their robustness to adversarial attacks has become an important topic for the community. Existing adversarial attacks directly add adversarial noise to the speech audio. However, prior work has shown that existing adversarial attacks face two limitations: they often transfer poorly to black-box ASR systems and are increasingly mitigated by defenses tailored to input-space perturbations. In this work, we propose a Clean-Referenced Feature-Vocoder Attack, a surrogate-based black-box attack that moves the adversarial search space from raw waveforms to self-supervised learning (SSL) representations. To address the transferability limitation, we perturb more generalizable acoustic-phonetic representations rather than low-level waveform samples, reducing dependence on surrogate-specific waveform gradients and encouraging adversarial perturbations that generalize across ASR systems. To bypass different defenses, we shift the adversarial signal from explicit additive waveform noise to SSL feature-space perturbations and reconstruct them through a vocoder into speech-like waveform adversarial signals, making the resulting samples less aligned with waveform-bounded defenses. Extensive experiments show that, when optimized only on raw Whisper-small as a public surrogate model, our attack transfers effectively to black-box ASR models with a +26.6 WER improvement over the SOTA baseline, while also remaining effective against multiple training defenses with a +36.2 WER improvement. These results reveal a blind spot in current ASR robustness evaluation.

Related papers

• Latent-Mark: An Audio Watermark Robust to Neural Resynthesis [62.09761127079914]
  Latent-Mark is the first zero-bit audio watermarking framework designed to survive semantic compression.<n>Our key insight is that robustness to the encode-decode process requires embedding the watermark within the invariant latent space.<n>Our work inspires future research into universal watermarking frameworks capable of maintaining integrity across increasingly complex and diverse generative distortions.
  arXiv  Detail & Related papers  (2026-03-05T15:51:09Z)
• Push-Pull: Characterizing the Adversarial Robustness for Audio-Visual Active Speaker Detection [88.74863771919445]
  We reveal the vulnerability of AVASD models under audio-only, visual-only, and audio-visual adversarial attacks. We also propose a novel audio-visual interaction loss (AVIL) for making attackers difficult to find feasible adversarial examples.
  arXiv  Detail & Related papers  (2022-10-03T08:10:12Z)
• Mel Frequency Spectral Domain Defenses against Adversarial Attacks on Speech Recognition Systems [33.21836814000979]
  This paper explores speech specific defenses using the mel spectral domain, and introduces a novel defense method called'mel domain noise flooding' (MDNF) MDNF applies additive noise to the mel spectrogram of a speech utterance prior to re-synthesising the audio signal. We test the defenses against strong white-box adversarial attacks such as projected gradient descent (PGD) and Carlini-Wagner (CW) attacks.
  arXiv  Detail & Related papers  (2022-03-29T06:58:26Z)
• Robustifying automatic speech recognition by extracting slowly varying features [16.74051650034954]
  We propose a defense mechanism against targeted adversarial attacks. We use hybrid ASR models trained on data pre-processed in such a way. Our model shows a performance on clean data similar to the baseline model, while being more than four times more robust.
  arXiv  Detail & Related papers  (2021-12-14T13:50:23Z)
• Blackbox Untargeted Adversarial Testing of Automatic Speech Recognition Systems [1.599072005190786]
  Speech recognition systems are prevalent in applications for voice navigation and voice control of domestic appliances. Deep neural networks (DNNs) have been shown to be susceptible to adversarial perturbations. To help test the correctness of ASRS, we propose techniques that automatically generate blackbox.
  arXiv  Detail & Related papers  (2021-12-03T10:21:47Z)
• Perceptual-based deep-learning denoiser as a defense against adversarial attacks on ASR systems [26.519207339530478]
  Adversarial attacks attempt to force misclassification by adding small perturbations to the original speech signal. We propose to counteract this by employing a neural-network based denoiser as a pre-processor in the ASR pipeline. We found that training the denoisier using a perceptually motivated loss function resulted in increased adversarial robustness.
  arXiv  Detail & Related papers  (2021-07-12T07:00:06Z)
• Towards Robust Speech-to-Text Adversarial Attack [78.5097679815944]
  This paper introduces a novel adversarial algorithm for attacking the state-of-the-art speech-to-text systems, namely DeepSpeech, Kaldi, and Lingvo. Our approach is based on developing an extension for the conventional distortion condition of the adversarial optimization formulation. Minimizing over this metric, which measures the discrepancies between original and adversarial samples' distributions, contributes to crafting signals very close to the subspace of legitimate speech recordings.
  arXiv  Detail & Related papers  (2021-03-15T01:51:41Z)
• WaveGuard: Understanding and Mitigating Audio Adversarial Examples [12.010555227327743]
  We introduce WaveGuard: a framework for detecting adversarial inputs crafted to attack ASR systems. Our framework incorporates audio transformation functions and analyses the ASR transcriptions of the original and transformed audio to detect adversarial inputs.
  arXiv  Detail & Related papers  (2021-03-04T21:44:37Z)
• WaveTransform: Crafting Adversarial Examples via Input Decomposition [69.01794414018603]
  We introduce WaveTransform', that creates adversarial noise corresponding to low-frequency and high-frequency subbands, separately (or in combination) Experiments show that the proposed attack is effective against the defense algorithm and is also transferable across CNNs.
  arXiv  Detail & Related papers  (2020-10-29T17:16:59Z)
• Class-Conditional Defense GAN Against End-to-End Speech Attacks [82.21746840893658]
  We propose a novel approach against end-to-end adversarial attacks developed to fool advanced speech-to-text systems such as DeepSpeech and Lingvo. Unlike conventional defense approaches, the proposed approach does not directly employ low-level transformations such as autoencoding a given input signal. Our defense-GAN considerably outperforms conventional defense algorithms in terms of word error rate and sentence level recognition accuracy.
  arXiv  Detail & Related papers  (2020-10-22T00:02:02Z)
• Characterizing Speech Adversarial Examples Using Self-Attention U-Net Enhancement [102.48582597586233]
  We present a U-Net based attention model, U-Net$_At$, to enhance adversarial speech signals. We conduct experiments on the automatic speech recognition (ASR) task with adversarial audio attacks.
  arXiv  Detail & Related papers  (2020-03-31T02:16:34Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.