Yang Xiang (项扬)
Hi~ This is Yang Xiang from Kunming, Yunnan, China, a spring city famous for its comfortable weather, beautiful scenery, and various delicious mushrooms.
Currently, I am an algorithm engineer at Hithink RoyalFlush AI Research Institute, Hangzhou, China. My research interests include speech enhancement, source separation, speech recognition, speech synthesis, deep representation learning (DRL), deep learning, and machine learning.
Previously, I was a Ph.D. student at Aalborg University and Capturi A/S, where I am working on DRL-based speech enhancement.
Email: xiangyang3131777@gmail.com
[CV (EN)] [CV(中文)] [Google Scholar] [LinkedIn]
Education
Aalborg University, Aalborg, Denmark
Ph.D. student in Electrical and Electronic Engineering, Sep. 2019 - Jun.2023.
Main supervisor: Prof. Mads Græsbøll Christensen
Co-supervisor: Dr. Morten Højfeldt Rasmussen and Dr. Jesper Lisby Højvang
Beijing University of Technology, Beijing, Chian
M.S. in Information and Communication Engineering, Sep. 2016 - Sep. 2019
Supervisor: Prof. Changchun Bao
B.E. in Electronic Information Engineering, Sep. 2012 - Sep. 2016
Teaching / Professional Experiences
Aalborg University, Aalborg, Denmark
Course lecture: Topics in Audio and Acoustic Signal Processing (2021), Nov. 2021
Project supervisor: Audio processing projects (2021), Feb. 2021 - Jun. 2021
Capturi A/S, Aarhus, Denmark
Industrial Ph.D. student, Sep. 2019 - Feb.2023
Working Experiences
Hithink RoyalFlush AI Research Institute, Hangzhou, China
Algorithm engineer, Jul. 2023 - present
1. Build a model-based speech enhancement module that supports two task objectives: improving auditory experience and enhancing speech recognition accuracy.
2. Participate in complex scenario speech recognition tasks, utilizing enhancement techniques to optimize solutions and improve recognition accuracy.
3. Explore cutting-edge technologies in speech enhancement, translating relevant research findings into patents or papers.
4. Participate in ICASSP 2024 ICMC-ASR challenge: response for front-end signal processing (Track 2: Rank 2).
Academic Services
Reviewer:
IEEE/ACM Trans. Audio,Speech, and Lang. Process;
ICASSP;
Speech Communication;
Applied Acoustics;
SCIENCE CHINA Information Sciences.
Selected Publications
Y. Xiang, J. Tian, X. Hu, X. Xu, Z. Yin, “A Deep Representation Learning-based Speech Enhancement Method Using Complex Convolution Recurrent Variational Autoencoder”, Accepted by ICASSP 2024.
Y. Xiang, J.L Højvang, M.H Rasmussen, and M.G Christensen, “A Two-stage Deep Representation Learning-based Speech Enhancement Method Using Variational Autoencoder and Adversarial Learning.” IEEE/ACM Transactions on Audio, Speech, and Language Processing 32 (2023): 164-177
S. Tao, Y. Xiang, H. Reddy, J.R Jensen, M.G Christensen, "Single Channel Speech Presence Probability Estimation based on Hybrid Global-Local Information." 2023 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). IEEE, 2023.
Y. Xiang, J.L Højvang, M.H Rasmussen, and M.G Christensen, “A deep representation learning speech enhancement method using β-VAE.” Accepted by Eurosipco 2022.
Y. Xiang, J.L Højvang, M.H Rasmussen, and M.G Christensen, “A Bayesian Permutation training deep representation learning method for speech enhancement with variational autoencoder.” in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process (ICASSP), 2022, pp. 381-385.
Y. Xiang, L. Shi, J.L Højvang, M.H Rasmussen, and M.G Christensen, “A Novel NMF-HMM Speech Enhancement Algorithm Based on Poisson Mixture Model” in Proc. IEEE Int. Conf. Acoust., Speech, Signal Process (ICASSP)., 2021, pp. 721-725.
Y. Xiang, L. Shi, J.L Højvang, M.H Rasmussen, and M.G Christensen, “An NMF-HMM speech enhancement method based on kullback-leibler divergence” in Proc. Interspeech,2020, pp. 2667–2671.
Y. Xiang and C. Bao, “A parallel-data-free speech enhancement method using multi-objective learning cycle-consistent generative adversarial network”, IEEE/ACM Trans. Audio, Speech, and Lang. Process., vol. 28, pp. 1826–1838, 2020.
Y. Xiang and C. Bao, “Speech enhancement via generative adversarial LSTM networks”, International Workshop on Acoustic Signal Enhancement (IWAENC), 2018, pp. 46-50.
Q. Huang, C. Bao, X. Wang, and Y. Xiang, “Speech enhancement method based on multi-band excitation model”, Applied Acoustics., vol. 163, pp. 10236, 2020.
Q. Huang, C. Bao, X. Wang, and Y. Xiang, “DNN-based speech enhancement using MBE model”, International Workshop on Acoustic Signal Enhancement (IWAENC), 2018, pp. 46-50.