MeanFlow-Accelerated Multimodal Video-to-Audio Synthesis via One-Step Generation

A key challenge in synthesizing audios from silent videos is the inherent trade-off between synthesis quality and inference efficiency in existing methods. Take Flow Matching based models as an example: they rely on modeling instantaneous velocity, inherently require an iterative sampling process, leading to slow inference speeds. To address this efficiency bottleneck, we introduce a MeanFlow-accelerated model that characterizes flow fields using average velocity, enabling one-step generation and thereby significantly accelerating multimodal video-to-audio (VTA) synthesis while preserving audio quality, semantic alignment, and temporal synchronization. Furthermore, a scalar rescaling mechanism is employed to balance conditional and unconditional predictions when classifier-free guidance (CFG) is applied, effectively mitigating CFG-induced distortions in one step generation. Since the audio synthesis network is jointly trained with multimodal conditions, we further evaluate it on text-to-audio (TTA) synthesis task. The results demonstrate that incorporating MeanFlow into the network significantly improves inference speed without compromising perceptual quality on both VTA and TTA tasks.