This is Yusong Wu’s home page. You can check out my CV, Publications and blogs.

Bio

2nd year graduate student of computer science in University of Montreal and Quebec Artificial Intelligence Institute (MILA) advised by Prof. Aaron Courville and Dr. Chengzhi Anna Huang.

Currently working on symbolic music generation, multi-modality learning.

Publications

  • Yusong Wu, Kun Chen, Ziyue Wang, Xuan Zhang, Fudong Nian, Xi Shao, Shengchen Li (2020). Audio Captioning Based on Transformer and Pre-Training for 2020 DCASE Audio Captioning Challenge. Technical Report, DCASE2020 Challenge.
  • Yusong Wu, Shengchen Li, Chenzhu Yu, Heng Lu, Chao Weng, Dong Yu (2020). Peking Opera Synthesis via Duration Informed Attention Network. INTERSPEECH 2020
  • Liqiang Zhang, Chengzhu Yu, Heng Lu, Chao Weng, Yusong Wu, Xiang Xie, Zijin Li, Dong Yu (2020). DurIAN-SC: Duration Informed Attention Network based Singing Voice Conversion System. INTERSPEECH 2020
  • Yusong Wu, Shengchen Li (2019). Guqin Dataset: A symbolic music dataset of Chinese Guqin collection . Proceedings of China Conference on Sound and Music Technology (CSMT 2019).
  • Yusong Wu, Shengchen Li (2019). Distinguishing Chinese Guqin and Western Baroque pieces based on statistical model analysis of melodies. 14th International Symposium on Computer Music Multidisciplinary Research (CMMR 2019).

Research interests

  • Automatic symbolic music composition
  • Multi-modality learning
  • Singing synthesis / Audio generation

Skills

  • Programming
    • C, C++, Python, Tensorflow, MATLAB.
  • Music
    • Over 15 years of experience playing percussion (proficient in timpani).
    • Over 10 years of experience in orchestra, wind orchestra and marching band.
    • Play with famous Chinese pop singer Jie Zhang (张杰) in 2016 on the show “Singer”. (我是歌手) Video

Blog

My blog on Zhihu.

News

26th July: Paper Peking Opera Synthesis via Duration Informed Attention Network and DurIAN-SC: Duration Informed Attention Network based Singing Voice Conversion System accepted by INTERSPEECH 2020.

1st July: won 2nd Place in DCASE 2020 Challenge Task 6: Automatic Audio Captioning.

8th November 2019: Presented the late-breaking/demo poster “Highly Expressive Peking Opera Synthesis with Durian System” on ISMIR 2019.

21st October 2019: Submitted a paper “Synthesising Expressiveness in Peking Opera via Duration Informed Attention Network” to ICASSP 2020 ( 45th International Conference on Acoustics, Speech, and Signal Processing). Please check out for audio samples

16th October 2019: Presented poster presentation “Distinguishing Chinese Guqin and Western Baroque pieces based on statistical model analysis of melodies” on CMMR 2019.

23th September 2019: (Accepted) Submitted an extended abstract “Highly Expressive Peking Opera Synthesis with Durian System” to late/breaking demo session at ISMIR 2019 (Proceedings of the 20th International Society for Music Information Retrieval Conference).

1st August 2019: Started internship at Tencent AI Lab.

15th July 2019: (Accepted) Submitted a paper “Guqin Dataset: A Symbolic Music Dataset” to CSMT2019 (China Conference on Sound and Music Technology).

9th May 2019: (Accepted) Submitted a paper “Distinguishing Chinese Guqin and Western Baroque pieces based on statistical model analysis of melodies” to CMMR 2019 (International Symposium on Computer Music Multidisciplinary Research).