Weidong Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Weidong Liu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Chen Guo
School of Electronics and Information Engineering, Jinggangshan University, No. 28 Xueyuan Road, Qingyuan District, Ji’an 343009, China
1 shared publication - Jingxiang Lv
Department of Computer Science, Jinggangshan University, China
1 shared publication - Pengjie Tang
Department of Computer Science, Jinggangshan University, China
1 shared publication - Shengbo Chen
School of Software, Nanchang University, Nanchang, China
1 shared publication - Ximin Liu
Shenzhen Ruier Electronics Co.,Ltd, China
1 shared publication
Latest Publications
- Open Access
- 26 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 335-341
Multi-source Fusion in Audio Scene Classification with Refined Transformers
With the rapid advancement of smart devices and Internet of Things (IoT) technologies, audio scene classification has become increasingly critical across various application domains. Deep learning models, particularly Convolutional Neural Networks (CNNs) and Transformer architectures, have garnered widespread attention for their exceptional performance in end-to-end audio classification tasks. Despite this, research on integrating multi-source information for more precise audio scene classification remains relatively scarce. In this paper, we introduce an audio classification model based on a Refined Transformer architecture, which effectively integrates environmental information from diverse sources, thereby improving the accuracy of recognition in complex audio environments. The model also achieves efficient processing of audio signals while maintaining low computational complexity. Experimental results on multiple public audio datasets demonstrate the model's effectiveness, with particularly enhanced robustness in noisy environments, showcasing its superior performance in audio scene classification tasks.
- 26 April 2024
- Published by Association for Computing Machinery (ACM)
- p. 335-341
Open AccessMulti-source Fusion in Audio Scene Classification with Refined Transformers
With the rapid advancement of smart devices and Internet of Things (IoT) technologies, audio scene classification has become increasingly critical across various application domains. Deep learning models, particularly Convolutional Neural Networks (CNNs) and Transformer architectures, have garnered widespread attention for their exceptional performance in end-to-end audio classification tasks. Despite this, research on integrating multi-source information for more precise audio scene classification remains relatively scarce. In this paper, we introduce an audio classification model based on a Refined Transformer architecture, which effectively integrates environmental information from diverse sources, thereby improving the accuracy of recognition in complex audio environments. The model also achieves efficient processing of audio signals while maintaining low computational complexity. Experimental results on multiple public audio datasets demonstrate the model's effectiveness, with particularly enhanced robustness in noisy environments, showcasing its superior performance in audio scene classification tasks.