Research map / Adversarial robustness and security / Attacks on text and graphs
VoiceBlock: Privacy through Real-Time Adversarial Attacks with Audio-to-Audio Models
Patrick O'Reilly, Andreas Bugler, Keshav Bhandari, Max Morrison, Bryan Pardo
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use VoiceBlock instead of universal additive perturbation | speaker de-identification · VoxCeleb1 quality | PESQ: 1.35 -> 3.74 | Improved |
| Use VoiceBlock instead of universal additive perturbation | speaker de-identification · VoxCeleb1 quality | STOI: 0.78 -> 0.92 | Improved |
| Use VoiceBlock instead of spectral gating | speaker de-identification · VoxCeleb1 quality | PESQ: 1.11 -> 3.74 | Improved |
| Use VoiceBlock instead of white noise | speaker de-identification · VoxCeleb1 quality | PESQ: 1.03 -> 3.74 | Improved |
| Use VoiceBlock instead of universal perturbation | ResNetSE34v2 (seen) · VoxCeleb1 | top-1 accuracy: 0.09 -> 0.03 | Mixed |
| Use VoiceBlock instead of spectral gating | ResNetSE34v2 (seen) · VoxCeleb1 | top-1 accuracy: 0.02 -> 0.03 | No clear effect |
| Use VoiceBlock instead of universal perturbation | Y-Vector (unseen) · VoxCeleb1 | top-1 accuracy: 0.48 -> 0.35 | Improved |
| Use spectral gating instead of VoiceBlock | Y-Vector (unseen) · VoxCeleb1 | top-1 accuracy: 0.35 -> 0.03 | Improved |
| Use white noise instead of VoiceBlock | Y-Vector (unseen) · VoxCeleb1 | top-1 accuracy: 0.35 -> 0.00 | Improved |
| Apply VoiceBlock vs no attack | ResNetSE34v2 · VoxCeleb1 | top-1 accuracy: 0.99 -> 0.03 | Improved |
| Apply VoiceBlock vs no attack | Y-Vector (unseen) · VoxCeleb1 | top-1 accuracy: 0.96 -> 0.35 | Improved |
| Use VoiceBlock instead of other attacks in listening test | MUSHRA listening study · 100 VoxCeleb1 clips | quality rating significance: p = 0.024 | Improved |
Closest papers by meaning
- AdvWave: Stealthy Adversarial Jailbreak Attack against Large Audio-Language Models (ICLR 2025 · similarity 0.86)
- I Can Hear You: Selective Robust Training for Deepfake Audio Detection (ICLR 2025 · similarity 0.85)
- Speed Master: Quick or Slow Play to Attack Speaker Recognition (AAAI 2025 · similarity 0.85)
- Silence is Golden: Leveraging Adversarial Examples to Nullify Audio Control in LDM-based Talking-Head Generation (CVPR 2025 · similarity 0.85)
- AudioMarkBench: Benchmarking Robustness of Audio Watermarking (NeurIPS 2024 · similarity 0.85)
- Adaptive Compression and Targeted Perturbation: A Unified Framework for Generalized Audio Deepfake Detection (NeurIPS 2026 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
