arrow
返回

Efficient audio-visual information fusion using encoding pace synchronization for Audio-Visual Speech Separation

delete2025-03-01
delete0
PRE
AI
X
Xinmeng Xu
W
Weiping Tu *
Y
Yuhong Yang *
DOI:10.1016/j.inffus.2024.102749delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Contemporary audio-visual speech separation (AVSS) models typically use encoders that merge audio and visual representations by concatenating them at a specific layer. This approach assumes that both modalities progress at the same pace and that information is adequately encoded at the chosen fusion layer. However, this assumption is often flawed due to inherent differences between the audio and visual modalities. In particular, the audio modality, being more directly tied to the final output (i.e., denoised speech), tends to converge faster than the visual modality. This discrepancy creates a persistent challenge in selecting the appropriate layer for fusion. To address this, we propose the Encoding Pace Synchronization Network (EPS-Net) for AVSS. EPS-Net allows for the independent encoding of the two modalities, enabling each to be processed at its own pace. At the same time, it establishes communication between the audio and visual modalities at corresponding encoding layers, progressively synchronizing their encoding speeds. This approach facilitates the gradual fusion of information while preserving the unique characteristics of each modality. The effectiveness of the proposed method has been validated through extensive experiments on the LRS2, LRS3, and VoxCeleb2 datasets, demonstrating superior performance over state-of-the-art methods.
Keyword:
Audio-visual speech separation
Encoding pace synchronization
Unimodal inter-layer synchronizer
Cross-modal intra-layer synchronizer

期刊

Information Fusion 封面图
Information Fusion
IF:
15.5
论文数:
4.2K
被引数:
2.7W

机构

W
wuhan university
学者数:
8.1W
论文数: 5.8W
被引数: 70
引用论文

引用论文

Solution structure of the Lewis x oligosaccharide determined by NMR spectroscopy and molecular dynamics simulations
err2002-05-01
err0
PREAI
errKristine E. Miller; Chaitali Mukhopadhyay; Perseveranda Cagas; C. Allen Bush
err分享
err收藏
Neonatal caffeine administration causes a permanent increase in the dendritic length of prefrontal cortical neurons of rats
err2006-08-04
err0
PREAI
errSergio Juárez‐méndez; Rosalba Carretero; Rubelia Martínez‐Tellez; Adriana B. Silva‐gómez; Gonzalo Flores
err分享
err收藏
Sprint Ability: How Well Does Your Software Exploit Bursts in Processing Capacity?
err2016-07-01
err0
PREAI
errNathaniel Morris; Siva Meenakshi Renganathan; Christopher Stewart; Robert Birke; Lydia Chen
err分享
err收藏
err分享
err收藏
Audiovisual Fusion: Challenges and New Approaches
err2015-09-01
err94
PREAI
errKatsaggelos, Aggelos K.; Bahaadini, Sara; Molina, Rafael
err分享
err收藏
err分享
err收藏
An Overview of Deep-Learning-Based Audio-Visual Speech Enhancement and Separation
err2021-01-01
err134
errOAAI
errMichelsanti, Daniel; Tan, Zheng-Hua; Zhang, Shi-Xiong; Xu, Yong; Yu, Meng; Yu, Dong; Jensen, Jesper
err分享
err收藏
Cost of Prescription Drug–Related Morbidity and Mortality
err2018-03-26
err0
errOAAI
errJonathan H. Watanabe; Terry McInnis; Jan D. Hirsch
err分享
err收藏
学者 查看更多内容