Feiteng Li
Full Stack Algorithm Engineer at Liulishuo
Yangpu District, Shanghai, China
Summary
🤩
Rockstar🎓
Top SchoolFeiteng Li is a Full Stack Algorithm Engineer with 12 years of experience specializing in speech recognition, speech assessment, and deep learning for production systems. Based in Shanghai, he has led CAPT and ASR efforts at Liulishuo, optimizing DNN/RNN acoustic models, speeding training across multi-GPU setups, and deploying lightweight models to mobile devices. He combines strong mathematical foundations in calculus, optimization, and statistics with hands-on engineering—implementing WFST decoders, VAD, and real-world TTS/ASR pipelines. An active open-source contributor, he reproduced and engineered a PyTorch VALL-E text-to-speech implementation, owning dataset pipelines and model architecture work. Known for turning algorithmic research into robust, performant products, he thrives at the intersection of math-heavy modeling and scalable engineering.
12 years of coding experience
3 years of employment as a software developer
MOOCBachelor's degree, Mathematics, Information and Computing Science, Bachelor's degree, Mathematics, Information and Computing Science at Dalian University of Technology
English
Github Skills (6)
pytorch10
machine-learning10
text-to-speech10
python10
data-processing10
large-language-models9
Programming languages (11)
Github contributions (5)
lifeiteng/vall-e
Jan 2023 - Mar 2023
PyTorch implementation of VALL-E(Zero-Shot Text-To-Speech), Reproduced Demo https://lifeiteng.github.io/valle/index.html
Role in this project:
Back-end Developer & ML Engineer
Back-end Developer & ML EngineerContributions:3 releases, 33 reviews, 137 commits in 1 month
Contributions summary:Feiteng's commits primarily focused on setting up and initializing the project, including the creation of the `setup.py` file for package management. Subsequent commits involved preparing the dataset, developing the necessary data processing pipeline, and defining the model architecture using PyTorch for a VALL-E model. These changes highlight the user's contributions to the core functionalities of the text-to-speech project.
chatgptin-context-learninglarge-language-modelstext-to-speechtts
lifeiteng/kaldi
Feb 2016 - Jul 2023
Contributions:2 PRs, 28 pushes, 6 branches in 7 years 5 months
kaldilocation
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.