Xubo Liu

Research Scientist, Meta Superintelligence Labs

Stony Stratford, England, United Kingdom
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts
email-iconphone-icongithub-logolinkedin-logotwitter-logostackoverflow-logofacebook-logo
Join Prog.AI to see contacts

Summary

🤩
Rockstar
🎓
Top School
Xubo Liu is a research scientist specializing in speech and audio foundation models, currently driving development of Meta Superintelligence Labs' "Her" with a focus on speech tokenization, full‑duplex modeling and post‑training for LLaMA4 speech systems. With a PhD from the University of Surrey and seven years' experience across Meta and Stability AI, he led Stability's speech R&D (including the Stable Codec work) and shipped the first production-ready full‑duplex voice assistant at Meta. He has co-authored over 80 papers across top AI and signal processing venues and serves as an area chair and frequent reviewer for flagship conferences. Xubo blends deep academic expertise with product-focused engineering, routinely moving research into demos and open-source artifacts, and he often tackles under-explored problems like token2wav and real-time codec-aware generation.
code7 years of coding experience
job1 year of employment as a software developer
bookDoctor of Philosophy - PhD, Electronic Engineering, Doctor of Philosophy - PhD, Electronic Engineering at University of Surrey
bookBachelor of Science - BS, Telecommunications Engineering, Bachelor of Science - BS, Telecommunications Engineering at Queen Mary University of London
github-logo-circle

Github Skills (13)

source-separation9
describe9
audio8
hosts8
nlp7
deep-learning5
pytorch5
attention-mechanism3
transformer2
text-to-speech2
synthesis2
speech2
speech-synthesis1

Programming languages (2)

HTMLPython

Github contributions (5)

github-logo-circle
personal homepage of Ziwei Liu
Contributions:42 commits, 100 pushes, 28 branches in 1 year 7 months
personal-homepage
liuxubo717/LASS

Mar 2022 - Jan 2023

This repo hosts the code and model of "Separate What You Describe: Language-Queried Audio Source Separation".
Contributions:24 commits, 1 PR, 20 pushes in 9 months
pytorchnlpsource-separationdescribedeep-learning
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.
Request Free Trial
Xubo Liu - Research Scientist, Meta Superintelligence Labs