Summary of Audiomarkbench: Benchmarking Robustness Of Audio Watermarking, by Hongbin Liu et al.
AudioMarkBench: Benchmarking Robustness of Audio Watermarking
by Hongbin Liu, Moyang Guo, Zhengyuan Jiang, Lun Wang, Neil Zhenqiang Gong
First submitted to arxiv on: 11 Jun 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Cryptography and Security (cs.CR); Sound (cs.SD); Audio and Speech Processing (eess.AS)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The increasing realism of synthetic speech, driven by advancements in text-to-speech models, raises ethical concerns regarding impersonation and disinformation. To address this issue, audio watermarking has emerged as a promising solution for embedding human-imperceptible watermarks into AI-generated audios. However, the robustness of audio watermarking against common and adversarial perturbations remains understudied. This paper presents AudioMarkBench, the first systematic benchmark for evaluating the robustness of audio watermarking against watermark removal and watermark forgery. The benchmark includes a new dataset created from Common-Voice across languages, biological sexes, and ages, three state-of-the-art watermarking methods, and 15 types of perturbations. The paper benchmarks the robustness of these methods against the perturbations in no-box, black-box, and white-box settings. The findings highlight the vulnerabilities of current watermarking techniques and emphasize the need for more robust and fair audio watermarking solutions. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Audio watermarking is a way to add secret information to AI-generated speech so it can be detected later. This helps prevent people from pretending to be someone else or spreading false information. Right now, there are concerns that synthetic speech is becoming too realistic, making it harder to tell what’s real and what’s fake. To address this issue, researchers created a new benchmark called AudioMarkBench. It includes a dataset of AI-generated audio, different methods for adding watermarks, and various types of disruptions or attacks that can try to remove the watermark. The goal is to find more robust and fair ways to add watermarks so they can’t be easily removed. |
Keywords
» Artificial intelligence » Embedding
