TTS-generated poisoned audio implants backdoors in SER models achieving high attack success at low poisoning ratios while preserving clean performance.
Fake the real: Back- door attack on deep speech classification via voice conversion,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Pmeta-TLA combines a frame-level timbre leakage trigger with meta-learning and PCGrad to inject multiple backdoors into speech models in one training run, claiming better attack success, stealth, and lower cost than baselines.
citing papers explorer
-
Backdoor Attacks on Speech Emotion Recognition via TTS-Generated Poisoning
TTS-generated poisoned audio implants backdoors in SER models achieving high attack success at low poisoning ratios while preserving clean performance.
-
Pmeta-TLA: Backdoor Attacks for Speech Classification Models via Meta-Learning with Timbre Leakage Attack
Pmeta-TLA combines a frame-level timbre leakage trigger with meta-learning and PCGrad to inject multiple backdoors into speech models in one training run, claiming better attack success, stealth, and lower cost than baselines.