SOURCE-LINKED INTELLIGENCE

Test-time adaptation for speech enhancement with an autoregressive speech prior

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Test-time adaptation (TTA) offers a promising direction for improving speech enhancement models under mismatched acoustic conditions, without requiring access to labeled target data. In this work, we propose a single-utterance TTA method that regularizes a pretrained speech enhancement model using an autoregressive prior trained on clean speech latent representations extracted from a neural audio codec. Adaptation is performed by minimizing the Kullback-Leibler divergence between the enhanced speech distribution and the clean speech prior. Experiments across multiple noisy speech datasets show

Read original source ↗ Open in workspace

recordType

paper

region

Global