MotionDuet: Dual-Conditioned 3D Human Motion Generation with Video-Regularized Text Learning Paper • 2511.18209 • Published Nov 22 • 1
MoCapAnything: Unified 3D Motion Capture for Arbitrary Skeletons from Monocular Videos Paper • 2512.10881 • Published 15 days ago • 29
Running on Zero Featured 86 Echo-TTS Preview 🐢 86 Fast, multi-speaker TTS (44.1kHz) with voice cloning