Speech-to-text freedom!
Combined NetEase's Confucius4-R2T2 streaming dictation with Jev to build a live caption tool that judges emotion and intent and annotates sentences in real time.
The original post, translated
Speech-to-text freedom! AI has marked all my emotions and intentions! I used NetEase Youdao's newly open-sourced Confucius4-R2T2 for streaming dictation, paired with TypeSafe's Jev for real-time judgment, and I built a "can-annotate" real-time subtitle tool that can recognize emotions and intentions 🎙️ Sentences that state key points → automatically bolded + hand-drawn underline. Sentences with emotion → automatically marked with emotion + background color.
Show the original post in its source language ▾
语音转写自由！AI 把我的情绪和意图全标出来了！ 我用网易有道刚开源的 Confucius4-R2T2 做流式听写，配上 TypeSafe 的 Jev 做实时判断，我搭了一个“会批注” 能 识别情绪和意图的实时字幕工具 🎙️ 说重点的句子 → 自动加粗＋手绘下划线 带情绪的句子 → 自动标情绪＋底色。
The post above is a machine translation from zh; the untranslated text is in the fold-out.
Engagement when collected
Numbers are a snapshot taken from X when the case was added to the library (schema v1, collected 2026-09-19); they will not match today.
Where this case fits
Filed under general Jev builds. In the pattern Jev is built for, the model answers a bounded question per step — and ordinary code acts on the answer, because the answer is already a value rather than a paragraph. Browse the full case library for the work it sits next to.
Related Jev cases
Keep browsing: all 1173 Jev cases · more from @limbopeng · builders · what Jev is
Last updated: 2026-09-22 · sources & corrections · every card links to its author's original post