EMAGE: Towards Unified Holistic Co-Speech Gesture Generation via Expressive Masked Audio Gesture Modeling

TL;DR

EMAGE is a novel framework that generates realistic, full-body co-speech gestures from audio, utilizing a new dataset and masked gesture modeling to improve gesture synthesis quality and diversity.

Contribution

It introduces BEAT2, a comprehensive 3D gesture dataset, and a Masked Audio Gesture Transformer for improved holistic gesture generation from speech.

Findings

State-of-the-art gesture generation performance

Effective integration of masked gesture priors

Flexible generation with predefined spatial-temporal inputs

Abstract

We propose EMAGE, a framework to generate full-body human gestures from audio and masked gestures, encompassing facial, local body, hands, and global movements. To achieve this, we first introduce BEAT2 (BEAT-SMPLX-FLAME), a new mesh-level holistic co-speech dataset. BEAT2 combines a MoShed SMPL-X body with FLAME head parameters and further refines the modeling of head, neck, and finger movements, offering a community-standardized, high-quality 3D motion captured dataset. EMAGE leverages masked body gesture priors during training to boost inference performance. It involves a Masked Audio Gesture Transformer, facilitating joint training on audio-to-gesture generation and masked gesture reconstruction to effectively encode audio and body gesture hints. Encoded body hints from masked gestures are then separately employed to generate facial and body movements. Moreover, EMAGE adaptively merges speech features from the audio's rhythm and content and utilizes four compositional VQ-VAEs to enhance the results' fidelity and diversity. Experiments demonstrate that EMAGE generates holistic gestures with state-of-the-art performance and is flexible in accepting predefined spatial-temporal gesture inputs, generating complete, audio-synchronized results. Our code and dataset are available https://pantomatrix.github.io/EMAGE/