Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | https://arxiv.org/abs/2503.11627 |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917331914981376 |
|---|---|
| author | Schwarzer, Will Chaudhari, Neel Thomas, Philip S. Fanelli, Andrea Liu, Xiaoyu |
| author_facet | Schwarzer, Will Chaudhari, Neel Thomas, Philip S. Fanelli, Andrea Liu, Xiaoyu |
| contents | Deep noise suppression (DNS) models enjoy widespread use throughout a variety of high-stakes speech applications. However, we show that four recent DNS models can each be reduced to outputting unintelligible gibberish through the addition of psychoacoustically hidden adversarial noise, even in low-background-noise and simulated over-the-air settings. For three of the models, a small transcription study with audio and multimedia experts confirms unintelligibility of the attacked audio; simultaneously, an ABX study shows that the adversarial noise is generally imperceptible, with some variance between participants and samples. While we also establish several negative results around targeted attacks and model transfer, our results nevertheless highlight the need for practical countermeasures before open-source DNS systems can be used in safety-critical applications. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_11627 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Are Deep Speech Denoising Models Robust to Adversarial Noise? Schwarzer, Will Chaudhari, Neel Thomas, Philip S. Fanelli, Andrea Liu, Xiaoyu Sound Machine Learning Audio and Speech Processing Deep noise suppression (DNS) models enjoy widespread use throughout a variety of high-stakes speech applications. However, we show that four recent DNS models can each be reduced to outputting unintelligible gibberish through the addition of psychoacoustically hidden adversarial noise, even in low-background-noise and simulated over-the-air settings. For three of the models, a small transcription study with audio and multimedia experts confirms unintelligibility of the attacked audio; simultaneously, an ABX study shows that the adversarial noise is generally imperceptible, with some variance between participants and samples. While we also establish several negative results around targeted attacks and model transfer, our results nevertheless highlight the need for practical countermeasures before open-source DNS systems can be used in safety-critical applications. |
| title | Are Deep Speech Denoising Models Robust to Adversarial Noise? |
| topic | Sound Machine Learning Audio and Speech Processing |
| url | https://arxiv.org/abs/2503.11627 |