Guochen Yu 余果宸
audio processing researcher at Kuaishou
Ph.D. Student at State Key Laboratory of Media Convergence and Communication, Communication University of China,
Beijing, China
Ph.D. Student at Key Laboratory of Noise and Vibration Research, Institute of Acoustics, Chinese Academy of Sciences,
Beijing, China
Email: yuguochen@kuaishou.com
Short Biography
I am currently a researcher at kuaishou, working on Large speech model and speech signal processing. I recieved my Ph.D. degree from Communication University of China and also jointly at Institute of Acoustics, Chinese Academy of Sciences (IACAS) supervised by Prof.Chengshi Zheng..
Before that, I received my bachelor's degree from the school of information and communication engineering at Communication University of China in 2017. I was a visiting student at mmlab of the school of information engineering at the Chinese University of Hong Kong (CUHK) in 2019. .
In my normal life, I like different kinds of sports, including basektball, swimming and Athletics. I used to be the captain of the basketball team of the Communication University of China, and led the team to achieve the top eight results in Beijing.
Research Interest
My research interests include large speech model (Zero-shot TTS and GPT-4O) and the front-end techniques for speech and audio, such as speech enhancement and speech signal improvement.
Publications and Manuscripts
2023
A General Deep Learning Speech Enhancement Framework Motivated by Taylor's Theorem
Andong Li, Guochen Yu, Chengshi Zheng, Wenzhe Liu, Xiaodong Li
IEEE/ACM Transactions on Audio, Speech, and Language Processing, (IEEE/ACM TASLP)
[Paper]
TaylorBeamixer: Learning Taylor-Inspired All-Neural Multi-Channel Speech Enhancement from Beam-Space Dictionary Perspective
Andong Li, Guochen Yu, Wenzhe Liu, Xiaodong Li, Chengshi Zheng
Submitted to Annual Conference of the International Speech Communication Association (Interspeech 2023)
[Paper]
2022
DBT-Net: Dual-branch federative magnitude and phase estimation with attention-in-attention transformer for monaural speech enhancement
Guochen Yu, Andong Li, Hui Wang, Yutian Wang, Yuxuan Ke, Chengshi Zheng
IEEE/ACM Transactions on Audio, Speech, and Language Processing, (IEEE/ACM TASLP)
[Paper]
[Demopage]
TMGAN-PLC: Audio Packet Loss Concealment using Temporal Memory Generative Adversarial Network
Yuansheng Guan, Guochen Yu, Andong Li, Chengshi Zheng, Jie Wang
Annual Conference of the International Speech Communication Association (Interspeech2022)
[Webpage]
Filtering and Refining: A Collaborative-Style Framework for Single-Channel Speech Enhancement
Andong Li, Guochen Yu, Chengshi Zheng, Xiaodong Li
IEEE/ACM Transactions on Audio, Speech, and Language Processing (IEEE/ACM TASLP)
[Code]
[Demopage]
Joint Magnitude Estimation and Phase Recovery Using Cycle-in-Cycle GAN for Non-Parallel Speech Enhancement
Guochen Yu, Andong Li, Yutian Wang, Chengshi Zheng, Hui Wang, Qin Zhang
in IEEE International Conference on Acoustics, Speech, & Signal Processing (ICASSP2022)
[Paper]
[Demopage]
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
Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA-ASC 2022)
[Paper]
Optimizing Shoulder to Shoulder: A Coordinated Sub-Band Fusion Model for Real-Time Full-Band Speech Enhancement
Guochen Yu, Andong Li, Wenzhe Liu, Chengshi Zheng, Yutian Wang, Hui Wang
International Symposium on Chinese Spoken Language Processing 2022 (ISCSLP 2022)
[Paper]
[Code]
[Demopage]
2021
CycleGAN-based Non-parallel Speech Enhancement with an Adaptive Attention-in-attention Mechanism
Guochen Yu, Yutian Wang, Chengshi Zheng, Hui Wang, Qin Zhang
in Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA-ASC 2021)
[Paper]
A Two-stage Complex Network using Cycle-consistent Generative Adversarial Networks for Speech Enhancement
Guochen Yu, Yutian Wang, Hui Wang, Qin Zhang, Chengshi Zheng,
Speech Communication, 2021
[Paper]
A Simultaneous Denoising and Dereverberation Framework with Target Decoupling
Andong Li, Wenzhe Liu, Xiaoxue Luo, Guochen Yu, Chengshi Zheng, Xiaodong Li
in Annual Conference of the International Speech Communication Association (Interspeech2021)
[Paper]
[Demopage]
2018-2020
Improved Relativistic Cycle-consistent GAN with Dilated Residual Network and Multi-Attention for Speech Enhancement
Yutian Wang*, Guochen Yu*, Jingling Wang, Hui Wang, Qin Zhang (*co-first author, student first author)
IEEE ACCESS, 2020
[Paper]
Selected Awards
[2021/03] INTERSPEECH 2022 Audio Deep Packet Loss Concealment Challenge (Rank 1 in non-causal methods. Best scores in PLCMOS, CMOS and DNSMOS). Check the website
[2022/01] ICASSP 2022 Deep Noise Suppression (DNS) Challenge in Track 1 non-personalized DNS. (Rank 9/27, Background noise suprression Rank 3). Check the website
[2021/03] Champion of INTERSPEECH 2021 Deep Noise Suppression (DNS) Challenge in 1st track. Check the website
[2020/10] Champion of ICASSP 2021 Deep Noise Suppression (DNS) Challenge in 1st track. Check the website
[2019/09] 'Honor Prize' of IEEE ISI-World Cup 2019 in Mission.1: Company Investment Value Evaluation