Rocket - Think it. Type it. Launch it.
Mini toolsDreamID-Omni
By Guoxu1233
DreamID-Omni unifies R2AV, RV2AV, and RA2V in one diffusion-transformer framework, using structured conditioning and disentanglement mechanisms designed to preserve identity and voice timbre while maintaining audio-video consistency in single- and multi-person scenarios.
New
Multimodal
Gen 3
Overview
DreamID-Omni is a human-centric audio-video generation framework that supports reference-conditioned generation, identity-and-voice video editing, and audio-driven identity animation with lip-sync.
Tools using DreamID-Omni
No tools found for this model yet.
Last updated: February 27, 2026
