AIConfucius R2T2 announced as an open-source streaming speech recognition model
A user said NetEase Youdao's Confucius R2T2 was open source on September 16, 2026, while the project's website listed model weights and code as “coming soon.”
TLDR
A user shared NetEase Youdao's Confucius R2T2 as an open-source release, linking to GitHub and the project website. The author highlighted domain-aware recognition, one stack for every latency target, and a design that treats context as a core input. The website described R2T2 as an open-source streaming speech recognition model, but listed its model weights and code as “coming soon.”
Combined views
659
1 Source, first seen 1d ago
Confucius R2T2 announced as an open-source streaming speech recognition model
A user said NetEase Youdao's Confucius R2T2 was open source on September 16, 2026, while the project's website listed model weights and code as “coming soon.”
TLDR
A user shared NetEase Youdao's Confucius R2T2 as an open-source release, linking to GitHub and the project website. The author highlighted domain-aware recognition, one stack for every latency target, and a design that treats context as a core input. The website described R2T2 as an open-source streaming speech recognition model, but listed its model weights and code as “coming soon.”