Yang (Austin) Xiao

——Do not go gentle into that good night.

About Me

Hello, I am Yang (Austin) Xiao, Xiao is my last name. Now I am an AI engineer at Fortemedia.

Recently I graduated as a CS Master student at NTU advised by Chng Eng Siong. I obtained my B.Eng degree from Jilin University. Then I worked as a software development engineer at China Mobile for one year.

I am contributing to building “easy-to-use”, “efficient”, and “robust” next-generation data-centric AI systems. At this moment, I mainly work in the data-centric speech learning system. Specifically, continual learning and domain adaptation apply in the speech domain to make the learning system continual, real-time, and efficient.

News

• [Dec. 2024] Invited to serve as a reviewer for ICME 2025.
• [Dec. 2024] Four papers have been accepted to ICASSP 2025 (One is the workshop).
• [Dec. 2024] Invited to serve as a reviewer for IJCNN 2025.
• [Sep. 2024] Invited to serve as a reviewer for ICASSP 2025.
• [Aug. 2024] Our paper about sound event detection has been accepted to DCASE Workshop 2024.
• [Feb. 2024] Our paper about sound event detection has been accepted to ICASSP 2024.
• [Sep. 2023] Invited to serve as a reviewer for ICASSP 2024.
• [May. 2023] Our paper about keyword spotting is accepted to INTERSPEECH 2023.
• [May. 2023] Achieved the 4th place on the DCASE 2023 Challenge on Task 4a: “Sound Event Detection with Weak Labels and Synthetic Soundscapes”.
• [Jan. 2023] Join the Fortemedia Singapore team as a software engineer.
• [Sep. 2022] Our paper about incremental learning and environmental sound classification is accepted to DCASE Workshop 2022.
• [Jun. 2022] Our paper about incremental learning and keyword spotting is accepted to INTERSPEECH 2022.

Research Interests

• Machine Learning System: Deep Learning Inference, Model Serving System, Edge-cloud Collaborative System
• Data-Centric AI: Incremental Learning, Active Learning, Few-shot Learning

Education

• Jilin University, Changchun, China
  • Aug.2016 - Jul.2020
  • B.E. in Internet of Things Engineering
  • Advisor: Professor Xuanjing Shen
• Peking University, Beijing, China
  • Jun.2020 - Aug.2020
  • Summer Term Enrollment
  • Advisor: Professor Hao Dan
• Nanyang Technological University, Singapore, Singapore
  • Aug.2021 - Jan.2023
  • Master of Science (Artificial Intelligence)
  • Advisor: Professor Chng Eng Siong

Professional Experience

• AI Engineer, Fortemedia Singapore
  • Jan.2023 - Present
• Software Engineer, China Mobile (Chengdu) Industrial Research Institute
  • Jul. 2020 - May.2021
  • Developed a smart cellular base station management system covering over 1.6 million stations in southeastern China.
• Student Participant, Summer 2021 of Open Source Promotion Plan
  • Jul. 2021 - Nov.2021
  • msflops: Count the MACs / FLOPs of your MindSpore model.
  • Developed a system to compute the theoretical amount of multiply-add operations in convolutional neural networks for MindSpore.
  • [Code]

Publications

• Leveraging LLM and Text-Queried Separation for Noise-Robust Sound Event Detection
  
  • Han Yin, Yang Xiao, Jisheng Bai, Rohan Kumar Das
    
  • Accepted to the IEEE ICASSP Workshop.2025.
    
  • [PDF] [Code]
• UCIL: An Unsupervised Class Incremental Learning Approach for Sound Event Detection
  
  • Yang Xiao,Rohan Kumar Das
    
  • Accepted to the IEEE ICASSP.2025.
    
  • [PDF]
• Dark Experience for Incremental Keyword Spotting
  
  • Tianyi Peng, Yang Xiao
    
  • Accepted to the IEEE ICASSP.2025.
    
  • [PDF]
• Exploring Text-Queried Sound Event Detection with Audio Source Separation
  
  • Han Yin, Jisheng Bai, Yang Xiao, Hui Wang, Siqi Zheng, Yafeng Chen, Rohan Kumar Das, Chong Deng, Jianfeng Chen
    
  • Accepted to the IEEE ICASSP.2025.
    
  • [PDF] [Code]
• WildDESED: An LLM-Powered Dataset for Wild Domestic Environment Sound Event Detection System
  
  • Yang Xiao,Rohan Kumar Das
    
  • Accepted to the DCASE Workshop.2024.
    
  • [PDF] [Project] [Code]
• Dual Knowledge Distillation for Efficient Sound Event Detection
  
  • Yang Xiao,Rohan Kumar Das
    
  • Accepted to the IEEE ICASSP Workshop.2024.
    
  • [PDF]
• Small Footprint Multi-channel Network for Keyword Spotting with Centroid Based Awareness
  
  • Dianwen Ng, Yang Xiao, Jia Qi Yip, Zhao Yang, Biao Tian, Qiang Fu, Eng Siong Chng, Bin Ma
    
  • Accepted to the 24rd INTERSPEECH Conference. 2023.
    
  • [PDF]
• Continual Learning For On-Device Environmental Sound Classification
  
  • Yang Xiao*, Xubo Liu*, James King, Arshdeep Singh, Eng Siong Chng, Mark D. Plumbley, Wenwu Wang
    
  • Accepted to the DCASE Workshop. 2022.
    
  • [PDF] [Code]
• Rainbow Keywords: Efficient Incremental Learning for Online Spoken Keyword Spotting
  
  • Yang Xiao, Nana Hou, Eng Siong Chng
    
  • Accepted to the 23rd INTERSPEECH Conference. 2022.
    
  • [PDF] [Code]

Services

• Reviewers: WASPAA, SPECOM, IALP, ICASSP, DCASE, IJCNN, ICME

Powered by Jekyll and Minimal Light theme.