School of Electronic and Computer Engineering,
Peking University, China

Email: dongchao98@stu.pku.edu.cn

Phone: +8615087581161

General

I am a third-year master student (the final year) at Peking University, majoring in Speech and Audio Processing. Before that, I received the Bachelor's Degree from Shanghai University in 2020.

My research focus on developing a human-agent that can communicate with human, e.g. understooding human's speech and environments sound, and then producing feedback to humans.
(Note: I plan to begin my PhD Research on The Chinese University of Hong Kong in 2023 fall, Supervised by Prof. Helen Meng. I am actively looking for any collaboration opportunities (e.g. NLP, speech synthesis, sound generation, speech/sound separation, sound event detection). Please feel free to contact me.)

Research Interests

Machine Learning, Audio Processing, Speech Processing

Educations

• School of Electronic and Computer Engineering, Peking University
  August 2020 - Now.

• School of Computer Engineering and Science, Shanghai University
  August 2016 - July 2020.

Experiences

• May 2021 - Now
  Tencent AI Lab, Speech Group, Intern.
  Supervisor: Chao Weng, Jianwei Yu and Bo Wu

• August 2020 - Now
  Peking University, ADSP Lab, master student.
  Supervisor: Yuexian Zou
  Co-author: Wenwu Wang

Publications

• Dongchao Yang, Helin Wang, Yuexian Zou
  Unsupervised Multi-Target Domain Adaptation for Acoustic Scene Classification
  Proc. Interspeech, 2021. [Code]

• Dongchao Yang, Helin Wang, Yuexian Zou, Zhongjie Ye, WenWu Wang
  A MUTUAL LEARNING FRAMEWORK FOR FEW-SHOT SOUND EVENT DETECTION
  ICASSP, 2022. [Code]

• Dongchao Yang, Helin Wang, Zhongjie Ye, Yuexian Zou
  Few-shot Bioacoustic Event Detection: A Good Transductive Inference is All You Need
  Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE), 2021. [Code]

• Dongchao Yang*, Helin Wang*, Zhongjie Ye, Yuexian Zou, WenWu Wang
  RaDur: A Reference-aware and Duration-robust Network for Target Sound Detection
  Proc. Interspeech, 2022. [Code]

• Dongchao Yang*, Helin Wang* , Chao Weng, Jianwei Yu, Yuexian Zou
  Improving Target Sound Extraction with Timestamp Information
  Proc. Interspeech, 2022. [Code]

• Yifei Xing, Dongchao Yang, Yuexian Zou
  Audio Pyramid Transformer with Domain Adaption for Weakly Supervised Sound Event Detection and Audio Classification
  Interspeech, 2022.

• Dongchao Yang*, Helin Wang* , Yujun Wang, Fan Cui, Yuexian Zou
  Detect What You Want: Target Sound Detection
  DCASE Workshop, 2022. [Code]

• Dongchao Yang*, Helin Wang, Yuexian Zou, WenWu Wang
  A Mixed Supervised Learning Framework for Target Sound Detection
  DCASE Workshop, 2022. [Code]

• Zhongjie Ye, Helin Wang, Dongchao Yang, Yuexian Zou
  Improving the Performance of Automated Audio Captioning via Integrating the Acoustic and Semantic Information
  DCASE2021 Workshop, 2021. [Code]

• Zhongjie Ye, Helin Wang, Dongchao Yang, Yuexian Zou
  Improving the Performance of Automated Audio Captioning via Integrating the Acoustic and Textual Information
  Challenge on Detection and Classification of Acoustic Scenes and Events (DCASE), 2021. [Code]

Under Review

• Dongchao Yang, Jianwei Yu, Helin Wang, Wen Wang, Chao Weng, Yuexian Zou, Dong Yu
  Diffsound: Discrete Diffusion Model for Text-to-sound generation
  TASLP, 2022. [Code]

Projects

• Member: Research on Deep Analysis Method of Acoustic Scenes for Smart Home Robot
  The project is a Shenzhen Science and Technology Fundamental Research Program, which studies the acoustic scenes and events in real home environments, including robust acoustic feature extraction, acoustic scene classification methods, abnormal sound event detection and warning.
  

• Member: Research on Multi-modal Health Monitoring System based on Infant Voices
  The project is a Shenzhen Science and Technology Fundamental Research Program, which studies the physiological characteristics of infant and conducts abnormal event detection based on audio and video signals.
  

Services

• IEEE/CAA Journal of Automatica Sinica
  Reviewer

Selected Awards