NetEase Youdao open-sources Confucius R2T2, a 2B speech model with 200ms real-time transcription

dr_cintas · x · 2026-09-17

NetEase Youdao released Confucius R2T2, a fully open-source 2B speech recognition model that transcribes in real time with 200ms latency. Its key feature: it commits each word the moment it's recognized and never rewrites it, making it well-suited for streaming use cases.

Related event: NetEase Youdao Open-Sources 2B Streaming ASR Model Confucius R2T2(15 posts)→

Original post →

More from Multimodal

Multimodal channel →