Guochen Yu
Communication University of China
16 Papers
26 Citations
Guochen Yu is an academic researcher from Communication University of China. The author has contributed to research in topics: Computer science & Engineering. The author has an hindex of 1, co-authored 1 publications.
Chat about Author
Papers
DBT-Net: Dual-Branch Federative Magnitude and Phase Estimation With Attention-in-Attention Transformer for Monaural Speech Enhancement
TL;DR: This paper proposes a dual-branch federative magnitude and phase estimation framework for monaural speech enhancement, aiming at recovering the coarse- and fine-grained regions of the overall spectrum in parallel, and employs a novel attention-in-attention transformer-based network within each branch for better feature learning.
Improved Relativistic Cycle-Consistent GAN With Dilated Residual Network and Multi-Attention for Speech Enhancement
TL;DR: Experimental results on a public dataset indicate that the proposed speech enhancement model achieves state-of-the-art speech enhancement performance, especially in reducing speech distortion and improving signal overall quality.
Filtering and Refining: A Collaborative-Style Framework for Single-Channel Speech Enhancement
TL;DR: This paper proposes a collaborative-style framework, namely, filtering and refining network (FRNet) for single-channel speech enhancement, recovering the complex spectrum of the target speech from coarse and fine-grained perspectives, and devise a two-branch structure dubbed filtering-refining module (FRM).
14
DMF-Net: A decoupling-style multi-band fusion model for real-time full-band speech enhancement
Guochen Yu,Yuansheng Guan,Weixin Meng,Chengshi Zheng,Hui Wang +4 more
TL;DR: A decoupling-style multi-band fusion model to perform full-band speech denoising and dereverberation that outperforms previous advanced systems and yields promising performance in terms of speech quality and intelligibility in real complex scenarios.
11
TaBE: Decoupling spatial and spectral processing with Taylor’s unfolding method in the beamspace domain for multi-channel speech enhancement
TL;DR: This paper proposes TaBE, a novel method that decouples spatial and spectral processing in multi-channel speech enhancement using Taylor's unfolding method in the beamspace domain, achieving competitive performance and providing insights into the enhancement pipeline design.
9