{"ID":2869288,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.14858","arxiv_id":"2509.14858","title":"MeanFlowSE: one-step generative speech enhancement via conditional mean flow","abstract":"Multistep inference is a bottleneck for real-time generative speech enhancement because flow- and diffusion-based systems learn an instantaneous velocity field and therefore rely on iterative ordinary differential equation (ODE) solvers. We introduce MeanFlowSE, a conditional generative model that learns the average velocity over finite intervals along a trajectory. Using a Jacobian-vector product (JVP) to instantiate the MeanFlow identity, we derive a local training objective that directly supervises finite-interval displacement while remaining consistent with the instantaneous-field constraint on the diagonal. At inference, MeanFlowSE performs single-step generation via a backward-in-time displacement, removing the need for multistep solvers; an optional few-step variant offers additional refinement. On VoiceBank-DEMAND, the single-step model achieves strong intelligibility, fidelity, and perceptual quality with substantially lower computational cost than multistep baselines. The method requires no knowledge distillation or external teachers, providing an efficient, high-fidelity framework for real-time generative speech enhancement. The proposed method is open-sourced at https://github.com/liduojia1/MeanFlowSE.","short_abstract":"Multistep inference is a bottleneck for real-time generative speech enhancement because flow- and diffusion-based systems learn an instantaneous velocity field and therefore rely on iterative ordinary differential equation (ODE) solvers. We introduce MeanFlowSE, a conditional generative model that learns the average ve...","url_abs":"https://arxiv.org/abs/2509.14858","url_pdf":"https://arxiv.org/pdf/2509.14858v3","authors":"[\"Duojia Li\",\"Shenghui Lu\",\"Hongchen Pan\",\"Zongyi Zhan\",\"Qingyang Hong\",\"Lin Li\"]","published":"2025-09-18T11:24:47Z","proceeding":"cs.SD","tasks":"[\"cs.SD\",\"cs.AI\"]","methods":"[\"Diffusion Model\"]","has_code":false,"code_links":[{"ID":609667,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2869288,"paper_url":"https://arxiv.org/abs/2509.14858","paper_title":"MeanFlowSE: one-step generative speech enhancement via conditional mean flow","repo_url":"https://github.com/liduojia1/MeanFlowSE","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}