# Exploiting Neural Audio Codec Latents for Adversarial Audio Attacks

Source: [Exploiting Neural Audio Codec Latents for Adversarial Audio Attacks](<https://www.sota2.com/research/paper/6a3a1315206542251d7c6fda>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-24T05:00:04.550Z

Authors: Sameek Bhattacharya, Bharath Krishnamurthy, Ajita Rattani

Published: 2026-06-18T19:40:46.000Z

## Abstract

Deep learning-based audio classification systems, including automatic speaker verification, are vulnerable to adversarial attacks. Realistic real-time threat assessment remains difficult because optimization-based methods, such as projected gradient descent (PGD) and Carlini-Wagner, require costly iterative updates in the high-dimensional waveform domain. Generative attacks allow single-shot synthesis but often introduce perceptible artifacts or depend on computationally intensive architectures, while diffusion and autoregressive approaches incur high inference latency. To address this gap, we propose a generative attack framework operating in the continuous latent space of a neural audio codec. A conditional generator synthesizes class-specific perturbations in a single forward pass and decodes them into adversarial waveforms. Our method achieves targeted attack success rates up to 99% with sub-7 ms inference, outperforming generative baselines while reducing latency by 24x.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Environmental Sound Classification on DCASE 2019](<https://www.sota2.com/research/sota/environmental-sound-classification-on-dcase-2019.md>): Environmental Sound Classification; DCASE 2019; Accuracy: 0.32.

- [Speaker Verification on LibriSpeech](<https://www.sota2.com/research/sota/speaker-verification-on-librispeech.md>): Speaker Verification; LibriSpeech; Accuracy: 0.02.

- [Speech Command Classification on Google Speech Commands v1 (test)](<https://www.sota2.com/research/sota/speech-command-classification-on-google-speech-commands-v1-test.md>): Speech Command Classification; Google Speech Commands v1 (test); Accuracy: 3.42.

- [Environmental Sound Classification on UrbanSound8k](<https://www.sota2.com/research/sota/environmental-sound-classification-on-urbansound8k.md>): Environmental Sound Classification; Urbansound8K; Accuracy: 1.23.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.20893>)

[PDF](<https://arxiv.org/pdf/2606.20893.pdf>)