Boosting ASR Robustness via Test-Time Reinforcement Learning with Audio-Text Semantic Rewards.

Linghan Fang, Tianxin Xie, Li Liu

Browse the full AAAI paper archive.