FLowHigh: Towards Efficient and High-Quality Audio Super-Resolution with Single-Step Flow Matching

TL;DR

FLowHigh introduces a single-step flow matching approach for efficient, high-quality audio super-resolution, significantly reducing latency while achieving state-of-the-art results on benchmark datasets.

Contribution

It integrates flow matching into audio super-resolution and develops probability paths tailored for high-resolution audio, enabling high-fidelity reconstruction with a single-step process.

Findings

Achieves state-of-the-art performance on VCTK dataset.

Maintains high audio quality with reduced computational cost.

Operates effectively with only one sampling step.

Abstract

Audio super-resolution is challenging owing to its ill-posed nature. Recently, the application of diffusion models in audio super-resolution has shown promising results in alleviating this challenge. However, diffusion-based models have limitations, primarily the necessity for numerous sampling steps, which causes significantly increased latency when synthesizing high-quality audio samples. In this paper, we propose FLowHigh, a novel approach that integrates flow matching, a highly efficient generative model, into audio super-resolution. We also explore probability paths specially tailored for audio super-resolution, which effectively capture high-resolution audio distributions, thereby enhancing reconstruction quality. The proposed method generates high-fidelity, high-resolution audio through a single-step sampling process across various input sampling rates. The experimental results on the VCTK benchmark dataset demonstrate that FLowHigh achieves state-of-the-art performance in audio super-resolution, as evaluated by log-spectral distance and ViSQOL while maintaining computational efficiency with only a single-step sampling process.