NetEase Youdao Open-Sources Confucius4-R2T2, a Streaming ASR That Never Rewrites Committed Text
rohanpaul_ai · x · 2026-09-18
NetEase Youdao open-sourced Confucius4-R2T2, a low-latency streaming ASR model with configurable 80ms–2s decoding chunks and an append-only output mode: once transcript text is committed, it is never revised. Rohan Paul highlights why this matters for agents — voice agents should consume speech incrementally but act only on committed text, since mutating transcripts corrupt downstream agent state. Targeted use cases include live captioning, NLP/LLM-agent pipelines, and simultaneous translation. The repo ships WebSocket server/client examples for quick deployment.
More from coding & agent
- Muse AI agent finds, calls and books a dentist in under 10 minutes, for free — ChrisUniverse · 2026-09-18
- Cursor Projects wins praise: kanban tracking, multi-model agents and 3-way bakeoffs — kieranklaassen · 2026-09-18
- OpenAI's Astra Struggles With Long-Term Architecture: Duplication, Overengineering, Bad Assumptions Linger — Substantial_Swan_144 · 2026-09-18
- Hang Ten Systems raises $53M more in seed funding, totaling $85M — vsikka · 2026-09-18
- Does AI coding actually become multiplayer? Shared Slack agents vs. supercharged solo devs — Telos_in_the_Void · 2026-09-18
- Skillry launches curated Agent Skills library to give Claude Code & Codex taste — yihui_indie · 2026-09-18