DyaDiT: A Multi-Modal Diffusion Transformer for Socially Favorable Dyadic Gesture Generation

TL;DR

DyaDiT is a multi-modal diffusion transformer that generates socially appropriate dyadic gestures by modeling interaction dynamics from paired audio signals, surpassing existing methods in quality and user preference.

Contribution

It introduces a novel multi-modal diffusion transformer that incorporates social context and mutual dynamics for dyadic gesture generation, with a motion dictionary and optional partner gesture modeling.

Findings

Outperforms existing methods on standard metrics

Users prefer DyaDiT-generated gestures over baselines

Demonstrates robustness and social appropriateness in gesture generation

Abstract

Generating realistic conversational gestures are essential for achieving natural, socially engaging interactions with digital humans. However, existing methods typically map a single audio stream to a single speaker's motion, without considering social context or modeling the mutual dynamics between two people engaging in conversation. We present DyaDiT, a multi-modal diffusion transformer that generates contextually appropriate human motion from dyadic audio signals. Trained on Seamless Interaction Dataset, DyaDiT takes dyadic audio with optional social-context tokens to produce context-appropriate motion. It fuses information from both speakers to capture interaction dynamics, uses a motion dictionary to encode motion priors, and can optionally utilize the conversational partner's gestures to produce more responsive motion. We evaluate DyaDiT on standard motion generation metrics and conduct quantitative user studies, demonstrating that it not only surpasses existing methods on objective metrics but is also strongly preferred by users, highlighting its robustness and socially favorable motion generation. Code and models will be released upon acceptance.