CogSR: Semantic-Aware Speech Super-Resolution via Chain-of-Thought Guided Flow Matching

TL;DR

CogSR is a novel speech super-resolution framework that uses semantic reasoning and acoustic priors to accurately restore severely degraded audio, surpassing existing models in fidelity and intelligibility.

Contribution

It introduces a Chain-of-Thought guided flow matching approach combined with semantic and acoustic priors for high-precision speech restoration.

Findings

Effectively eliminates ambiguity in severely degraded audio

Restores high-frequency details with linguistic accuracy

Robustly improves speech quality in legacy and surveillance recordings

Abstract

Applying speech super-resolution (SR) to recordings with severely low sampling rates is a critical challenge in digital archiving and investigative audio recovery. In these scenarios, the input lacks essential acoustic cues. Consequently, existing generative models often fail; without sufficient context, they hallucinate phonetic content, guessing words based on probability rather than meaning. To address this, we propose CogSR, a framework designed specifically for high-precision, offline restoration. Our approach shifts the focus from simple signal mapping to cognitive reconstruction. By integrating a Large Audio-Language Model, we employ Chain-of-Thought reasoning to act as a semantic anchor, while explicit acoustic priors ensure the speaker's identity remains consistent. This guides a Rectified Flow backbone to synthesize high-frequency details that are not only realistic but linguistically accurate. Evaluations show that CogSR effectively eliminates ambiguity in severe degradation regimes, making it a robust solution for restoring high-value legacy and surveillance audio.