Szu-Wei Fu

Make everything as simple as possible, but not simpler

Meteora!     This picture was taken in Greece when I attended SLT, 2018.



About Me

I received my Ph.D. from the CS department at National Taiwan University in 2020, under the advisement of Prof. Shou-De Lin.
and Dr. Yu Tsao.

My research interests include: Machine Learning, Deep learning, Speech enhancement and processing.


News (Last Update:2021/06)

  • 2021/06:   Join Microsoft as data and applied scientist!
  • 2021/06:   Two papers are accepted in Interspeech 2021 ( MetricGAN+ and Phone-Fortified Perceptual Loss )
  • 2020/05:   Join the SpeechBrain project at Mila, Canada. Advisor: Prof. Yoshua Bengio and Dr. Mirco Ravanelli
  • 2020/01:   Successfully defense my Ph.D. thesis
  • 2020/01:   One paper is accepted in IEEE/ACM TASLP 2020
  • 2019/11:   Two papers are accepted in IEEE Signal Processing Letters (SPL) 2019
  • 2019/07:   Start my Research Intern at Starkey Hearing Technologies, MN, USA. Advisor: Dr. Zhang Tao
  • 2019/07:   Three papers are accepted in Interspeech 2019
  • 2019/05:   One paper is accepted in ICML 2019 (with Travel Awards)
  • 2018/09:   One paper is accepted in SLT 2018
  • 2018/06:   One paper is accepted in Interspeech 2018 (with Travel Grants)
  • 2018/03:   One paper is accepted in IEEE/ACM TASLP
  • 2017/10:   One paper is accepted in IEEE APSIPA 2017
  • 2017/08:   One paper is accepted in IEEE MLSP
  • 2016/12:   One paper is accepted in IEEE TBME
  • 2016/10:   Start my Research Intern at NICT, Kyoto, Japan. Advisor: Dr. Xugang Lu
  • 2016/06:   One paper is accepted in Interspeech 2016
  • 2016/03:   One paper is accepted in Entropy

Recent Publications

For a complete publication list, please see my CV

MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancement

ICML, 2019 (with Travel Awards)
paper
code

Learning with Learned Loss Function: Speech Enhancement with Quality-Net to Improve Perceptual Evaluation of Speech Quality

IEEE Signal Processing Letters (SPL) , 2019
paper

Quality-Net: An End-to-End Non-intrusive Speech Quality Assessment Model based on BLSTM

Interspeech, 2018 (with Travel Grants)
paper
code

End-to-End Waveform Utterance Enhancement for Direct Evaluation Metrics Optimization by Fully Convolutional Neural Networks

IEEE/ACM Transactions on Audio, Speech, and Language Processing, 2018
paper
code

Raw Waveform-based Speech Enhancement by Fully Convolutional Networks

IEEE APSIPA, 2017
paper

Complex Spectrogram Enhancement by Convolutional Neural Network with Multi-Metrics Learning

IEEE International Workshop on Machine Learning for Signal Processing, 2017
paper

Joint Dictionary Learning-based Non-Negative Matrix Factorization for Voice Conversion to Improve Speech Intelligibility After Oral Surgery

IEEE Transactions on Biomedical Engineering, 2016
paper
code

SNR-Aware Convolutional Neural Network Modeling for Speech Enhancement

Interspeech, 2016
paper

Contact me:

If you are interested in my research, please contact me.