DiM-Gesture: Co-Speech Gesture Generation with Adaptive Layer Normalization Mamba-2 framework

Speech-driven gesture generation is an emerging domain within virtual human creation, where current methods predominantly utilize Transformer-based architectures that necessitate extensive memory and are characterized by slow inference speeds. In response to these limitations, we propose \textit{DiM...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Zhang, Fan, Ji, Naye, Gao, Fuxing, Zhao, Bozuo, Wu, Jingmei, Jiang, Yanbing, Du, Hui, Ye, Zhenqing, Zhu, Jiayang, Zhong, WeiFan, Leyao Yan, Ma, Xiaomeng
Format Paper
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 01.08.2024
Subjects
Diffusion layers
Diffusion rate
Feature extraction
Fuzzy sets
Gesture recognition
Inference
Speech
State-of-the-art reviews
Synchronism
Virtual humans
Virtual memory systems
Online AccessGet full text

Cover

More Information
Summary:Speech-driven gesture generation is an emerging domain within virtual human creation, where current methods predominantly utilize Transformer-based architectures that necessitate extensive memory and are characterized by slow inference speeds. In response to these limitations, we propose \textit{DiM-Gestures}, a novel end-to-end generative model crafted to create highly personalized 3D full-body gestures solely from raw speech audio, employing Mamba-based architectures. This model integrates a Mamba-based fuzzy feature extractor with a non-autoregressive Adaptive Layer Normalization (AdaLN) Mamba-2 diffusion architecture. The extractor, leveraging a Mamba framework and a WavLM pre-trained model, autonomously derives implicit, continuous fuzzy features, which are then unified into a singular latent feature. This feature is processed by the AdaLN Mamba-2, which implements a uniform conditional mechanism across all tokens to robustly model the interplay between the fuzzy features and the resultant gesture sequence. This innovative approach guarantees high fidelity in gesture-speech synchronization while maintaining the naturalness of the gestures. Employing a diffusion model for training and inference, our framework has undergone extensive subjective and objective evaluations on the ZEGGS and BEAT datasets. These assessments substantiate our model's enhanced performance relative to contemporary state-of-the-art methods, demonstrating competitive outcomes with the DiTs architecture (Persona-Gestors) while optimizing memory usage and accelerating inference speed.
ISSN:2331-8422