CogSR: Semantic-Aware Speech Super-Resolution via Chain-of-Thought Guided Flow Matching
Jiajun Yuan, Xiaochen Wang, Yuhang Xiao, Yulin Wu, Chenhao Hu, Xueyang Lv
TL;DR
CogSR tackles the challenge of restoring severely degraded speech by reframing SR as cognitive reconstruction. It fuses a Diffusion Transformer with Rectified Flow in a latent DAC space, guided by a Large Audio-Language Model via structured Chain-of-Thought and explicit acoustic priors on bandwidth and pitch. The dual guidance constrains semantic content and spectral fidelity, reducing hallucinations and preserving speaker identity. Extensive experiments show state-of-the-art results in intelligibility, spectral accuracy, and perceptual quality, demonstrating strong potential for legacy and investigative audio restoration.
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.