SongBsAb: A Dual Prevention Approach against Singing Voice Conversion based Illegal Song Covers
Guangke Chen
Network and Distributed System Security (NDSS) Symposium 2025 · Day 1 · Audio Security
In an era increasingly dominated by AI-generated content, the music industry faces unprecedented challenges, particularly from **AI-based automated song covers**. These sophisticated tools leverage **Singing Voice Conversion (SVC)** technology to transform a song's vocal rendition from one singer's style and timbre to another's, all while meticulously preserving the original lyrics and melody. While SVC holds potential for beneficial applications, such as fans creating personalized content, its widespread accessibility through open-source toolkits has led to a surge in illegal song covers, infringing upon artists' rights and record companies' copyrights. Guangke Chen's presentation introduces **SongBsAb**, a pioneering dual prevention approach designed to proactively combat these unauthorized AI-generated covers.
AI review
Solid applied ML security research that adapts adversarial perturbation techniques to a genuinely underserved problem — proactive poisoning of SVC pipelines before a song ships. The dual-objective framing (identity + lyric disruption) and the psychoacoustic masking refinement are real contributions, but the work sits squarely in the adversarial-examples-for-audio lineage and doesn't break new ground methodologically. Competent, publishable, conference-filler quality.