d3ro-voice/docs
Yun Chan 3cac14f668
Some checks failed
ci / 정본·보안·린트·타입·테스트 (push) Has been cancelled
deploy-site / deploy (push) Has been cancelled
ci / 워크스페이스 빌드 검증 (push) Has been cancelled
ci / 모바일 린트·타입·Jest (push) Has been cancelled
ci / Supabase Edge Functions + Cloudflare Worker (push) Has been cancelled
ci / .NET API 서버 테스트 (push) Has been cancelled
feat(desktop): typo auto-correct and context-aware sentence prediction
Typo auto-correct (INPUT-19)
- TypoCorrectionService + core typo-correction: 2-set Korean layout
  mistypes both ways, always-wrong spellings, near-misses of the user's
  frequent words, optional model check re-validated in core; applied by
  backspace N + paste and cancelled on new keys or window change.

Prediction quality (INPUT-20/21)
- gemma models are sent raw with the user's unfinished sentence pre-filled
  into the model turn, so the model truly continues the text (mid-word
  joins and spacing exact). Other models keep the instruction prompt.
- Context: text before the caret (opening lines and a fact list of names,
  dates, amounts outside the 400-char window), messages just sent in the
  same chat window (memory only, 30 min), personal-graph follow-ups and
  IDF-ranked related sentences with Korean stemming.
- Guards: first sentence only, restated-prefix stripping (also fixes the
  instruction path), repeat drop, no invented dates/names/numbers
  (one retry), plain -다 register, no emoji unless the user uses them,
  search-box keywords. Fill loop mixes instruction prompts for variety.
- Opt-in real-model probe: tests/integration/suggestion-quality.ollama.test.ts

Suggestions now actually show up
- Terminal (Claude Code): caret trust from the input line, not the whole
  screen (status line below the box blocked every keystroke); Korean
  composition counts as typing at the end; a grey placeholder filling the
  empty box at once is not user text.
- Chrome: UIA focus flicker while typing no longer resets "edited".
- KakaoTalk/RichEdit: trailing paragraph mark ignored for edit position
  and stripped so accepted text keeps its separating space.
- While the overlay is visible the suggestion shortcuts are captured so
  Ctrl+Alt+Enter no longer also types a newline; released when hidden.
- nut-js preloaded after boot (first accept waited 4.8 s).
2026-10-03 21:48:28 +09:00
..
deployment docs: record Wave 3 surface consolidation and fold NAS-only whisper into compose 2026-09-26 16:02:44 +09:00
design feat(bootstrap): Whisper large-v3-turbo 기본 전환 + 온보딩 2단계 다운로드 진행률 2026-07-21 11:59:49 +09:00
design-refs 초기 프로젝트 설정: 하네스 시스템 + 설계서 + RE 노하우 2026-04-05 01:03:03 +09:00
evidence feat: add mobile shell and unified search 2026-09-28 23:23:36 +09:00
map feat(desktop): typo auto-correct and context-aware sentence prediction 2026-10-03 21:48:28 +09:00
phases docs: record Wave 3 surface consolidation and fold NAS-only whisper into compose 2026-09-26 16:02:44 +09:00
re-findings Phase 8: SSOT 리팩토링 + HotkeyRecordModal + Dashboard 재작성 2026-04-05 09:54:55 +09:00
tdd-red docs: record the 1.1.0 release and add the infrastructure map 2026-09-16 23:27:52 +09:00
v2 docs: record Wave 3 surface consolidation and fold NAS-only whisper into compose 2026-09-26 16:02:44 +09:00
v3 feat: add mobile shell and unified search 2026-09-28 23:23:36 +09:00
monetization-plan.md refactor(billing): remove Stripe; payments are Payple (web) and Google Play (mobile) 2026-09-26 20:56:18 +09:00
REFACTOR_POLICY.md refactor(billing): remove Stripe; payments are Payple (web) and Google Play (mobile) 2026-09-26 20:56:18 +09:00
REFACTOR_WAVE1_REPORT.md docs: refactor-wave Wave 1 정책·보고서·현황 (메타) 2026-07-22 02:01:26 +09:00
REFACTOR_WAVE2_REPORT.md docs: refactor-wave Wave 2 보고서·현황 (메타) 2026-07-22 02:40:08 +09:00
REFACTOR_WAVE3_REPORT.md docs: record Wave 3 surface consolidation and fold NAS-only whisper into compose 2026-09-26 16:02:44 +09:00