An uncertainty-aware Audio-LLM with confidence-based attention modulation reduces whispered-speech CER by 17% relative and hallucination rate from over 25% to 4.5%.
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
CONDITIONAL 1representative citing papers
citing papers explorer
-
Whisper-Aware LLM: Self-Supervised Uncertainty Learning for Robust Whispered Speech Recognition
An uncertainty-aware Audio-LLM with confidence-based attention modulation reduces whispered-speech CER by 17% relative and hallucination rate from over 25% to 4.5%.