Grounded Decoding for Autoregressive Speech Enhancement via Adaptive Code-Space Grounding and Local LLM Refinement
An evidence-grounded generative SE framework that uses a deterministic estimate as imperfect evidence as imperfect evidence is proposed and SNR-Conditioned CSG (SNR-CSG), which maps a calibrated residual-SNR estimate to an utterance-level strength and constructs an adaptive grounded anchor is introduced.