Jie Pu
Hi, I am a research scientist at Zoom, working on multilingual speech recognition and language modelling.
My research works cover the general topic of AI speech and language, which includes speech recognition (ASR), speech synthesis (TTS), speaker recognition, speaker diarization and speech separation.
Before joining Zoom, I was working at the University of Cambridge with Prof. Mark Gales. I received a PhD degree from Imperial College London, under the supervision of Prof. Maja Pantic and Dr. Yannis Panagakis.
News
2022 - 10: Join Zoom AI team as a research scientist.
2021 - 04: Join the Speech Research Group of Machine Intelligence Lab at Cambridge, as a research associate.
2020 - 08: Join the Amazon Alexa science team at California, USA for an applied scientist internship.
2020 - 05: Our paper gets accepted at the IEEE Signal Processing Letters.
2020 - 04: Join the Machine Learning team in ARM research for a four-month internship.
2018 - 11: Our paper gets accepted to the IEEE Transactions on Cybernetics.
2017 - 02: Our ICASSP 2017 paper is selected as a finalist for the Student Paper Contest.
Research Topics and Selected Publications
- Automatic Speech Recognition (ASR):
Multi-stage Large Language Model Correction for Speech Recognition [Paper]
Jie Pu, Thai-Son Nguyen, Sebastian Stüker
- Speaker Recognition:
Scaling Effect of Self-supervised Speech Models [Paper]
Jie Pu, Yuguang Yang, Ruirui Li, Oguz Elibol, Jasha Droppo
- Speech Synthesis (Text-to-Speech, TTS):
Building Synthetic Speaker Profiles in Text-to-speech Systems [Paper]
Jie Pu, Yixiong Meng, Oguz Elibol
- Speech Separation:
Blind Audio–Visual Localization and Separation via Low-Rank and Sparsity [Paper]
Jie Pu, Yannis Panagakis, Stavros Petridis, Jie Shen, Maja Pantic
- Speaker Diarization:
Multimodal Video Search by Examples (UK EPSRC funded) [Project]
Research project with Universities of Cambridge, Surrey and Ulster and the BBC