MeanSE: Efficient Generative Speech Enhancement with Mean Flows
Abstract
Speech enhancement (SE) improves degraded speech's quality, with generative models like flow matching gaining attention for their outstanding perceptual quality. However, the flow-based model requires multiple numbers of function evaluations (NFEs) to achieve stable and satisfactory performance, leading to high computational load and poor 1-NFE performance. In this paper, we propose MeanSE, an efficient generative speech enhancement model using mean flows, which models the average velocity field to achieve high-quality 1-NFE enhancement. Experimental results demonstrate that our proposed MeanSE significantly outperforms the flow matching baseline with a single NFE, exhibiting extremely better out-of-domain generalization capabilities.
Keywords
Cite
@article{arxiv.2509.21214,
title = {MeanSE: Efficient Generative Speech Enhancement with Mean Flows},
author = {Jiahe Wang and Hongyu Wang and Wei Wang and Lei Yang and Chenda Li and Wangyou Zhang and Lufen Tan and Yanmin Qian},
journal= {arXiv preprint arXiv:2509.21214},
year = {2025}
}
Comments
Submitted to ICASSP 2026