Wenzhe Liu

     

I currently work at Kuaishou, and was employed by Tencent, and focus on speech/audio signal processing with deep learning for Real-Time Communication, and audio generation. I am interested in audio codec, speech front-end processing, Text-to-Speech, and microphone array processing.

(may need to enter ctrl(command)+shift+R to refresh the page)

Research Interests

RTC front-end processing (3A): (general) speech enhancement, echo cancallation

audio compression (codec): audio coding (including speech, music, and noise), BWE, and PLC

generative AI for speech: text-to-speech, voice conversion, and LLMs Applied to TTS

microphone array processing: beamforming, Neural beamformer, and multi-speaker DOA estimation

Education
Institute of Acoustics, Chinese Academy of Sciences (中科院声学所)
M.S. in Signal and Information Processing
Sept 2019 - Jun 2022
Harbin Engineering University (哈尔滨工程大学)
Bachelor of Underwater Acoustics Engineering
Sept 2015 - Jun 2019
Experiences
Audio Algorithm Engineer
Kuaishou Technology (快手), Beijing
Supervisor: Chen Zhang
Nov 2023 - Now
Applied Researcher
Tencent Ethereal Audio Lab (腾讯天籁实验室), Beijing & Shenzhen
Supervisor: Yuepeng Li
Jul 2022 - Nov 2023
Applied Research Intern
Tencent Ethereal Audio Lab, Shenzhen
Jul 2020 - Aug 2020
Reviewer
IEEE Transactions on Audio, Speech and Language Processing
International Conference on Acoustics, Speech, and Signal Processing (ICASSP)
The International Speech Communication Association (ISCA) reviewer (InterSpeech)
IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA) 2023
IEEE Open Journal of Signal Processing
Mobile Networks and Applications