Chao Yang
Technical Development Manager at Horizon Robotics
Suzhou City, Jiangsu, China
Summary
🤩
Rockstar🎓
Top SchoolChao Yang is a Technical Development Manager with 10 years of experience building production speech and language systems, currently leading engineering efforts at Horizon Robotics from Suzhou. He brings deep hands-on expertise in Python/C++, Linux, NLP, speech recognition and ML, backed by a Master's in CS/Machine Learning. Previously he led speech teams at Mobvoi—co-authoring an industry-oriented open-source end-to-end ASR toolkit (WeNet) and shipping popular Mandarin TTS tools—demonstrating both product and open-source impact. His GitHub contributions include adding torchaudio support and removing Kaldi dependencies in WeNet, showing a focus on making research-grade models production-ready. Known for bridging algorithm research and scalable engineering, he also has experience developing dialogue, NLU, and knowledge-graph services from his Microsoft and AISpeech tenures. Colleagues describe him as a pragmatic leader who prioritizes robust pipelines and measurable deployment outcomes.
10 years of coding experience
8 years of employment as a software developer
Bachelor, Computer Science, 4.5/5, Top4 in 200 student, Bachelor, Computer Science, 4.5/5, Top4 in 200 student at Northwestern Polytechnical UniversityMaster's degree, Computer Science, Master's degree, Computer Science at Northwestern Polytechnic University (CN)
Chinese, English
Github Skills (9)
pytorch10
machine-learning10
speech-recognition10
automatic-speech-recognition10
data-processing9
python9
transformer9
asr8
conformer8
Programming languages (7)
Github contributions (5)
wenet-e2e/wenet
Dec 2020 - May 2022
Production First and Production Ready End-to-End Speech Recognition Toolkit
Role in this project:
ML Engineer
ML EngineerContributions:42 reviews, 25 commits, 26 PRs in 1 year 5 months
Contributions summary:Chao made significant contributions to the Wenet toolkit, a production-ready end-to-end speech recognition system, by implementing support for torchaudio. This involved adding torchaudio dataloaders, integrating wav distortion techniques, and refactoring existing Kaldi dependencies. The user also refined scripts and comments and fixed related bugs, demonstrating a strong understanding of speech recognition pipelines and audio processing techniques.
pytorchend-to-endasrproduction-readyspeech-to-text
placebokkk/placebokkk.github.io
Aug 2020 - Apr 2022
My blog
Contributions:30 commits, 86 pushes in 1 year 8 months
my-blog
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.