TimeSteer: Inference-Time Speech Scheduling in Joint Audio-Visual Diffusion Models
A training-free framework that localizes each utterance's source span through Source Span Localization and transfers the associated audio-visual latent content from the source interval to the specified target interval through Region-Aware Latent Remapping is proposed.
Chao Zhou, Yilu Chen, Qi Chu et al.
· 0 citations