Xiang Hao (郝翔)

Fourth-year Ph.D. student, Department of Data Science and Artificial Intelligence, The Hong Kong Polytechnic University.

• Email: haoxiangsnr at gmail.com
• Links: Google Scholar / GitHub / LinkedIn

Research Interests

Hearing-aid signal processing, speech enhancement, target speaker extraction, beamforming, auditory attention decoding, neuromorphic audio, and real-time edge deployment.

Employment

• Research Intern, Tencent AI Lab, Shenzhen, China, Mar. 2023 - Mar. 2024.
  Mentored by Dr. Jianwei Yu. Worked on text-guided target speaker extraction for multi-talker scenarios.

• Research Intern, Tencent Ethereal Audio Lab, Shenzhen, China, Dec. 2020 - Jun. 2021.
  Mentored by Dr. Yannan Wang. Worked on neural beamforming and real-time speech enhancement for communication scenarios.

• Visiting Student, Westlake University, Hangzhou, China, Jun. 2020 - Dec. 2020.
  Mentored by Prof. Xiaofei Li. Developed FullSubNet, a full-band and sub-band fusion model for speech enhancement.

• Research Intern, Sogou Inc., Beijing, China, Jul. 2019 - Oct. 2019 and Feb. 2020 - Jun. 2020.
  Worked on speech enhancement and target speaker extraction, including a two-stage masking and inpainting approach for low-SNR conditions.

• Development Intern, Meituan, Beijing, China, Mar. 2016 - Sept. 2016.
  Developed front-end data visualization components with React and Node.js.

Education

• The Hong Kong Polytechnic University, Ph.D. Candidate in Data Science and Artificial Intelligence, May 2023 - Present.
  Focus: ultra-low-power hearing-aid signal processing. Supervisors: Prof. Jibin Wu and Prof. Kay Chen Tan.

• The Chinese University of Hong Kong, Ph.D. Student in Systems Engineering and Engineering Management, Sept. 2021 - Nov. 2022.
  Focus: environmentally robust automatic speech recognition. Supervisor: Prof. Xunying Liu.

• Inner Mongolia University, M.Sc. in Computer Science, Sept. 2018 - Jun. 2021.
  Thesis: Multi-channel speech enhancement using constrained optimization and deep learning. Supervisor: Prof. Xiangdong Su.

Selected Awards

• Winner, The Clarity Enhancement Challenge CEC3 Task 1, Dec. 2024.
• Best Paper Award, IEEE Conference on Artificial Intelligence, Jun. 2024.
• Winner, Intel Neuromorphic Deep Noise Suppression Challenge, Track 1, Nov. 2023.
• Outstanding Research Postgraduate Thesis, Inner Mongolia Autonomous Region, Jul. 2021.
• National Scholarship, Ministry of Education, China, Oct. 2020.
• Elite Member, Tencent Rhino-Bird Elite Training Program, Apr. 2020.

Selected Publications

For the full list, please see Publications. My name is bolded.

1. Xiang Hao, Jibin Wu, Jianwei Yu, Chenglin Xu, and Kay Chen Tan, “Typing to Listen at the Cocktail Party: Text-Guided Target Speaker Extraction,” IEEE Transactions on Cognitive and Developmental Systems, 2025. DOI

2. Xiang Hao, Chenxiang Ma, Qu Yang, Jibin Wu, and Kay Chen Tan, “Toward Ultralow-Power Neuromorphic Speech Enhancement With Spiking-FullSubNet,” IEEE Transactions on Neural Networks and Learning Systems, 2025. DOI / Code

3. Xiang Hao, Chenxiang Ma, Qu Yang, Kay Chen Tan, and Jibin Wu, “When Audio Denoising Meets Spiking Neural Network,” IEEE Conference on Artificial Intelligence, 2024. Best Paper Award. DOI

4. Xiang Hao, Xiangdong Su, Radu Horaud, and Xiaofei Li, “FullSubNet: A Full-Band and Sub-Band Fusion Model for Real-Time Single-Channel Speech Enhancement,” ICASSP, 2021. DOI / Code

5. Xiang Hao, Shixue Wen, Xiangdong Su, Yun Liu, Guanglai Gao, and Xiaofei Li, “Sub-Band Knowledge Distillation Framework for Speech Enhancement,” INTERSPEECH, 2020. DOI

6. Xiang Hao, Xiangdong Su, Shixue Wen, Zhiyu Wang, Yiqian Pan, Feilong Bao, and Wei Chen, “Masking and Inpainting: A Two-Stage Speech Enhancement Approach for Low SNR and Non-Stationary Noise,” ICASSP, 2020. DOI

Invited Talks

• Sonova (Shanghai), “Deep Learning for Hearing Aids,” remote, Dec. 2024.
• Intel Labs, “Toward Next-Generation Neuromorphic Audio Denoising with Spiking-FullSubNet,” remote, Dec. 2023.

Skills

• Programming: Python, PyTorch, JavaScript, C/C++, MATLAB.
• Languages: Chinese, English.