AudioGuard combines waveform-level and policy-based detection to improve safety in audio AI systems, outperforming baselines on a new benchmark spanning languages, voices, and threat types.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
AudioGuard: Toward Comprehensive Audio Safety Protection Across Diverse Threat Models
AudioGuard combines waveform-level and policy-based detection to improve safety in audio AI systems, outperforming baselines on a new benchmark spanning languages, voices, and threat types.