MeanFlowSE: one-step generative speech enhancement via conditional mean flow
This addresses the bottleneck of multistep inference for real-time speech enhancement, offering an efficient solution for applications requiring low-latency processing.
The paper tackles the problem of real-time generative speech enhancement by introducing MeanFlowSE, a model that learns average velocity over finite intervals to enable single-step inference, achieving strong intelligibility, fidelity, and perceptual quality on VoiceBank-DEMAND with substantially lower computational cost than multistep baselines.
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.