PhysWave: Physics-Guided Latent Diffusion Models for Controllable Spatial Audio Generation
PhysWave unifies natural-language and trajectory control through a shared waypoint-caption representation, and augments diffusion training with two differentiable acoustic priors: spherical-harmonic direction consistency and inverse-square distance consistency that help generate spatially consistent FOA audio while mai...
Ling-Feng Yao, Chenpei Huang, Xing-Ke Yang et al.
· 0 citations