• v1.17.1 6c6875bbee

    D3RO Voice v1.17.1
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 50s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 40s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 28s
    ci / .NET API 서버 테스트 (push) Successful in 1m0s
    deploy-site / deploy (push) Successful in 42s
    release / release-windows (push) Failing after 7m2s
    portable-unsigned / portable-windows (push) Failing after 13m39s
    Stable

    yunchan released this 2026-11-14 02:59:49 +00:00 | 9 commits to main since this release

    [1.17.1] - 2026-11-14

    Supersedes unpublished 1.17.0 with the full-length ASAR Pickle header smoke verification fix and ships the full 1.17.0 multi-surface release.

    Fixed

    • Read the full declared JSON byte length (up to 32 MiB) instead of a fixed 64 KiB prefix in verify-desktop-installer-smoke.mjs when validating resources/app.asar.

    SHA-256

    • D3RO-Voice-Setup-1.17.1-x64.exe: 15718b9d7770ecc0a2472856667feb65d158070542bc8cc0826f03eef2c8d75f
    • D3RO-Voice-Setup-1.17.1-x64.exe.blockmap: 2fd2373cc876a0d2ac41c2a6e55e563c4b31ec64600adfa6c462d64823820905
    • latest.yml: 1088e10de3ac0780db3e4f96e3fd119c8ca5746f271fd15217f919af354d862e
    • latest.yml.sig: a14dbfbe10abdc30f2030824d6697665a9220c2e9b488a937a407ee3f6ef47a0
    Downloads
  • v1.16.5 f57e40d82a

    D3RO Voice v1.16.5
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 20s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 46s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 20s
    ci / .NET API 서버 테스트 (push) Successful in 13s
    deploy-site / deploy (push) Successful in 33s
    portable-unsigned / portable-windows (push) Failing after 5m12s
    release / release-windows (push) Successful in 13s
    Stable

    yunchan released this 2026-11-06 14:25:00 +00:00 | 18 commits to main since this release

    [1.16.5] - 2026-11-06

    Transcript polish now guards what the model must not break, measured against the local model.

    Changed

    • Polish instructions were rewritten to the wording measured best on the local model: fillers may only be removed from a fixed list, meaning qualifiers such as 거의/대부분/아직 must stay, a word that makes the sentence fall apart (for example 유치역학 → 유체역학) should be repaired, and speaker labels or timestamps that are not in the input must not be invented.
    • Comma-separated fragments from speech are joined into sentences, and one prompt is shared by the app, the tests and the verification probe.

    Fixed

    • The model could add a [00:00] or [화자 1] marker that was never in the audio; these are now stripped from the saved transcript, so timestamps and speakers come only from recognition and diarization.
    • The model sometimes deleted meaning-bearing qualifiers (for example turning "사실 거의 근간" into "사실 근간"). Qualifiers the speaker actually said are restored from the original chunk.
    • Unknown-word repair stays model-dependent (measured 2 of 3 runs on the same clip); once a term is in your dictionary or has been typed a few times, the fix becomes deterministic.

    SHA-256

    • D3RO-Voice-Setup-1.16.5-x64.exe: 025f492948787c3303a20f93119cf211dae618f007bf0f8c3720a7308d36cfdf
    • D3RO-Voice-Setup-1.16.5-x64.exe.blockmap: 9474ca1e6cd38bca56863a178d7b2fe2c3e4e08147427f8eaeb8a999e7a05671
    • latest.yml: 627b46be77c2274751d3cbd433fa1c14e2f4d98ce4e753aac7ae808c66a09aa2
    • latest.yml.sig: 05e29077385b327a61edc1391388f243af1e3ce700206d54abd0d664290d7044
    Downloads
  • v1.16.4 7dd6178dcf

    D3RO Voice v1.16.4
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 25s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 50s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 27s
    ci / .NET API 서버 테스트 (push) Successful in 14s
    deploy-site / deploy (push) Successful in 49s
    portable-unsigned / portable-windows (push) Failing after 5m39s
    release / release-windows (push) Failing after 11m50s
    Stable

    yunchan released this 2026-11-06 08:23:22 +00:00 | 24 commits to main since this release

    [1.16.4] - 2026-11-06

    AI correction of transcripts is now grounded in the vocabulary the app learned from typing.

    Changed

    • AI polishing (recorded or pasted transcripts) treats learned vocabulary as the correction source: names, product names and in-house terms the transcript misheard are rewritten to the learned spelling, while ordinary nouns and endings are left alone. The learned term list and the user expressions are given to the model, and dictionary pronunciations plus typed identifiers are fixed deterministically before and after the model runs.
    • Dictation output is checked against the learned vocabulary once more after the language model finishes, so a term the model rewrote is restored.

    Fixed

    • Hangul-layout typing artifacts (dkssudgktpdy) no longer appear in the term list, the correction hints, or the style examples handed to the model.

    SHA-256

    • D3RO-Voice-Setup-1.16.4-x64.exe: 857d957e68084b9e41d2644fdd47d7b9bbab933f651201d4776dea84b1980b93
    • D3RO-Voice-Setup-1.16.4-x64.exe.blockmap: 41122bd8e54445b1e4c0b967ac486aefd81502f0c6f2feac1fb366541593525c
    • latest.yml: ee444ddde70da2bfb06522a8b9d6f414186f40119d3b1a3eb2678867dd431feb
    • latest.yml.sig: b1cf1327f2f08052fe6146a15ef813f2ba90629d9837b30ffa3e401c7d14966e
    Downloads
  • v1.16.3 1f5372bfcc

    D3RO Voice v1.16.3
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 44s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 47s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 28s
    ci / .NET API 서버 테스트 (push) Successful in 18s
    deploy-site / deploy (push) Successful in 47s
    portable-unsigned / portable-windows (push) Failing after 36m55s
    release / release-windows (push) Successful in 38s
    Stable

    yunchan released this 2026-11-06 04:40:45 +00:00 | 26 commits to main since this release

    [1.16.3] - 2026-11-06

    Supersedes 1.16.2, which could rewrite ordinary Korean words in transcripts.

    Fixed

    • Stop the learned-vocabulary pass from rewriting ordinary Korean. Measured against real meeting transcripts it changed 286 of 3,005 words (하면→라면, 이제→이게, 매일→메일, 정보→정도). Deterministic correction is now limited to your dictionary's pronunciations and the Latin identifiers you type; Korean terms reach recognition through the prompt hint and the polish prompt instead. Re-measured: 0 of 3,005 words changed.
    • Drop Hangul-layout typing artifacts (for example dkssudgktpdy) from the recognition hint so they no longer bias transcription, and keep the original casing of learned identifiers (VSCode).

    Changed

    • Korean terms are no longer auto-replaced word by word. They are used as a recognition hint and as reference for the local model, which decides with sentence context.

    SHA-256

    • D3RO-Voice-Setup-1.16.3-x64.exe: b0bd6500494c53b209c079d7d9e78e3e3cd87b8f425112086d15fe86acdfab3e
    • D3RO-Voice-Setup-1.16.3-x64.exe.blockmap: 1e3667982a858bc54ddc2e981bdc0005ccb89fc834169b3b5756e0fd23d18328
    • latest.yml: 320150610a6d28256afd285a29550ef32172dee9f25a2b37b8e6b37d380e8438
    • latest.yml.sig: 7ab2c8d8269af8c868c76fea1d38fb5b103f0bf705a73e7018dbec5614f7b4f2
    Downloads
  • v1.16.2 a21de78303

    D3RO Voice v1.16.2
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 1m6s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 48s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 24s
    ci / .NET API 서버 테스트 (push) Successful in 20s
    deploy-site / deploy (push) Successful in 54s
    portable-unsigned / portable-windows (push) Failing after 14m51s
    release / release-windows (push) Successful in 19m48s
    Stable

    yunchan released this 2026-11-05 12:59:50 +00:00 | 28 commits to main since this release

    [1.16.2] - 2026-11-05

    Published from an annotated tag. Installer and update metadata are served by the
    canonical Forgejo feed; no binaries are committed to this repository.

    Added

    • Meeting and live-caption transcription now use the vocabulary the app learned from your typing: dictionary words, pronunciations and frequently typed terms bias speech recognition before it runs, fix mistyped words as soon as each line is finalized (no model required), and are given to the local model when it polishes a line.
    • Polished caption lines now reach the meeting transcript and the meeting screen instead of stopping inside the caption service, so the saved transcript carries the corrected text.

    Fixed

    • Learned vocabulary used to feed typing suggestions only, which is why transcripts kept mangling names and jargon the app already knew.

    SHA-256

    • D3RO-Voice-Setup-1.16.2-x64.exe: 30ab53548de28511e2459f7173dfdf86a80d3505ad9feb1e6dff1d905b2f1337
    • D3RO-Voice-Setup-1.16.2-x64.exe.blockmap: 840c78fe4a4685cc265bdb37d3877a7e88a0017f447a5b574a7f3136056e5761
    • latest.yml: f72f87be839f9b4823169419a83b8397b0d06c7c84220628d3d54cb6e85b0530
    • latest.yml.sig: d65427f523cc8bdf7766f0594f33c89ae40f4895cadc8edd6a0a393415aa0f1d
    Downloads
  • v1.16.1 77e6796595

    D3RO Voice v1.16.1
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 45s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 52s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 31s
    ci / .NET API 서버 테스트 (push) Successful in 19s
    deploy-site / deploy (push) Successful in 34s
    release / release-windows (push) Failing after 14m12s
    portable-unsigned / portable-windows (push) Failing after 14m15s
    Stable

    yunchan released this 2026-11-03 22:06:32 +00:00 | 32 commits to main since this release

    [1.16.1] - 2026-11-04

    Published from an annotated tag. Installer and update metadata are served by the
    canonical Forgejo feed; no binaries are committed to this repository.

    Fixed

    • prevent fast correction input races

    SHA-256

    • D3RO-Voice-Setup-1.16.1-x64.exe: bf5843a0ed53b0a054046a54cb608e643a24ebb4f02f6a3d2cdefeed82e9be8d
    • D3RO-Voice-Setup-1.16.1-x64.exe.blockmap: 8e5d252d0365829bd895817d70c08617cf8aa8dc7b6c9a4a2adc4c2c815dc363
    • latest.yml: 9d12b2c0127741b6b7d08eb913a597a0d039f4f0b9c79e1383c3804818763ae7
    • latest.yml.sig: b3591294031cb5537bc37df27717dde33b3c3148f69c587ca8c23179fc2865e0
    Downloads
  • v1.16.0 7f9ec45868

    D3RO Voice v1.16.0
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 34s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 56s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 39s
    ci / .NET API 서버 테스트 (push) Successful in 22s
    deploy-site / deploy (push) Successful in 42s
    release / release-windows (push) Failing after 10m17s
    portable-unsigned / portable-windows (push) Failing after 10m24s
    Stable

    yunchan released this 2026-11-01 10:39:47 +00:00 | 35 commits to main since this release

    [1.16.0] - 2026-11-01

    Added

    • Import existing audio or video recordings into a meeting, retaining timestamped transcripts and playable audio for minutes generation.

    Fixed

    • Cancel meeting imports when the account changes, preventing transcripts from being saved to another account.
    • Deliver the updated local speech engine with CPU thread limits and faster terminal UIA handling. Existing engines upgrade before use.

    Changed

    • Release the required speech engine before publishing the desktop update, and verify the packaged engine starts successfully.

    SHA-256

    • D3RO-Voice-Setup-1.16.0-x64.exe: e9596ee25fe03e4ebf34c526198d2f44ddc7cdd90ec354eb304216687486a7c0
    • D3RO-Voice-Setup-1.16.0-x64.exe.blockmap: c93149498ea1a6b33ae49c1bea34779a4c6f385d7063f6550a436c8a54acdefa
    • latest.yml: 671b3a6004addf335e9f667b330e40999361c79f09826872d4e4d19738a9b5cf
    • latest.yml.sig: 2c633bcbb093ae33dcbeca74f724ec6557ec5d6e19f78a2f2161da18e0c61ce0
    Downloads
  • v1.15.4 5a8c5cd792

    D3RO Voice v1.15.4
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 43s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 50s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 29s
    ci / .NET API 서버 테스트 (push) Successful in 10s
    deploy-site / deploy (push) Successful in 50s
    release / release-windows (push) Successful in 11m18s
    portable-unsigned / portable-windows (push) Failing after 14m51s
    Stable

    yunchan released this 2026-10-31 14:15:11 +00:00 | 37 commits to main since this release

    [1.15.4] - 2026-10-31

    Published from an annotated tag. Installer and update metadata are served by the
    canonical Forgejo feed; no binaries are committed to this repository.

    Changed

    • Suggestion generation pipeline modularization (SuggestionService).
      • Separates model-request construction, first-candidate retries, generation-session lifecycle, stale-result validation, failure fallback, and filled-candidate ranking/publication into named boundaries while preserving cancellation, timeout, personal-memory fallback, and candidate ordering behavior.

    Fixed

    • Desktop test reliability and offline release gate.
      • Restores the 35 previously failing offline tests by preserving real STT exports in partial mocks, consolidating popup renderer recovery coverage, and aligning built-in instruction and synthetic Unicode input expectations with current behavior.
      • Splits Vitest into explicit unit, service, integration, evaluation, and all scopes; the default desktop test gate now runs 2,764 deterministic offline tests with zero failures.

    SHA-256

    • D3RO-Voice-Setup-1.15.4-x64.exe: 7b75b6ea8fa8c88c6db28dd14afd2de3fe0f490095ace0e66d7ec756158a03c1
    • D3RO-Voice-Setup-1.15.4-x64.exe.blockmap: f11254a95c8e5d7c3eb97e9bf381b92e1716a91a68506c7255d3c6fdbfee9925
    • latest.yml: b1f12083aff52bac5549ed6f462250ea067abfc67598fd75223042a9201fb2ca
    • latest.yml.sig: 0d729024b20106221568fb40d428aab10bc2125cc419c8e4fb413d9052a2e58f
    Downloads
  • v1.15.3 fe0989c456

    D3RO Voice v1.15.3
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 21s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 48s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 27s
    ci / .NET API 서버 테스트 (push) Successful in 15s
    deploy-site / deploy (push) Successful in 25s
    release / release-windows (push) Failing after 13m41s
    portable-unsigned / portable-windows (push) Failing after 13m46s
    Stable

    yunchan released this 2026-10-30 19:46:38 +00:00 | 39 commits to main since this release

    [1.15.3] - 2026-10-31

    Published from an annotated tag. Installer and update metadata are served by the
    canonical Forgejo feed; no binaries are committed to this repository.

    Fixed

    • Ctrl + Alt + Enter suggestion acceptance reliability (@d3ro/core/keybinding-runtime, KeyBindingService, SuggestionService, bootstrap).
      • Fixes ChordStateMachine.keyUp so releasing a shared modifier (Ctrl or RightAlt) releases all held chords in _isDown that depend on that modifier instead of stopping after the first match (command Ctrl+RightAlt), preventing suggestion-accept (Ctrl+Alt+Enter) from remaining stuck in _isDown and silently dropping subsequent presses as auto-repeat.
      • Maps UiohookKey.NumpadEnter and numpad navigation keys in UIOHOOK_TO_VK and adds a 25 ms OS globalShortcut fallback trigger in KeyBindingService._syncSuggestionCapture so Ctrl+Alt+Enter still fires even if uiohook misses a key event.
      • Keeps OS suggestion shortcut capture active while the "생성 중..." pending indicator or an accept is in flight (isOverlayOrCaptureActive), and when Ctrl+Alt+Enter is pressed while candidate #1 is still streaming (_inFlight / _generating / _pending), waits up to 2.5 s for candidate #1 to finish and immediately accepts it instead of ignoring the key press.
    • Duplicate typo-model check elimination (TypoCorrectionService.propose).
      • Skips re-running _askModel in TypoCorrectionService.propose when handleTypingContext has already verified the same region (_isRegionChecked(region)), preventing a redundant 336-token LLM call during suggestion candidate filling from evicting Ollama's suggestion KV cache.

    SHA-256

    • D3RO-Voice-Setup-1.15.3-x64.exe: 43b8ed12a2ca1df877a53dedb02f48db0a1567e0255f30bc3dc052082d887623
    • D3RO-Voice-Setup-1.15.3-x64.exe.blockmap: 6202cc1b8ecaed8caca7af4f5b41dccc7eae9902d52aa265d609de421b9b2b5a
    • latest.yml: e6c649612b4da1faef32923af56723b2027c36d5aec6f5ee4c5f25e2f17c062b
    • latest.yml.sig: 81ee39d271c84eeb840e9f4e171e90d20c9964b13ea1297d9523ffd841a374d8
    Downloads
  • v1.15.1 6d909dd0e5

    D3RO Voice v1.15.1
    Some checks failed
    ci / 정본·보안·린트·타입·테스트 (push) Failing after 30s
    ci / 워크스페이스 빌드 검증 (push) Has been skipped
    ci / 모바일 린트·타입·Jest (push) Successful in 41s
    ci / Supabase Edge Functions + Cloudflare Worker (push) Successful in 25s
    ci / .NET API 서버 테스트 (push) Successful in 10s
    deploy-site / deploy (push) Successful in 29s
    portable-unsigned / portable-windows (push) Failing after 3m59s
    release / release-windows (push) Successful in 15s
    Stable

    yunchan released this 2026-10-30 18:39:54 +00:00 | 43 commits to main since this release

    [1.15.1] - 2026-10-31

    Published from an annotated tag. Installer and update metadata are served by the
    canonical Forgejo feed; no binaries are committed to this repository.

    Fixed

    • Terminal & xterm.js Korean IME character-swallowing fix (터미널·에이전트 창 한글 타이핑 시 1글자씩 씹히는 현상 근본 해결).
      • Removes synchronous cross-process isForegroundKoreanIme (SendMessageTimeoutW(WM_IME_CONTROL)) calls from the low-level keydown hook (InputTelemetryService._handleKeyDown) that were disrupting Windows TSF/DefIME composition state on every keystroke in terminal windows, replacing them with a one-time idle-snapshot IME sync (TerminalKeystrokeTracker.syncKoreanIme).
      • Adds a dedicated xterm-helper-textarea fast path in sidecar/uia_bridge.py that skips GetTopLevelControl() DOM tree walks and TextPattern.GetSelection() / ExpandToEnclosingUnit(Character) / TextEditPattern.GetActiveComposition() COM calls (which stalled Chromium's renderer thread up to 1,500 ms during Korean IME composition and left isComposing stuck true), and extends _WAKE_TTL_SECONDS from 60 s to 86,400 s per HWND.
      • Suppresses 40 ms word-boundary UIA snapshots on normal Korean words (_shouldRunFastTerminalBoundarySnapshot), removes the duplicate 680 ms _typoTimer, and stops 800 ms _sampleWhileTyping polling on terminal surfaces so UIA never queries xterm-helper-textarea while you press the first consonant of the next word.
      • Prevents TypoCorrectionService from injecting commitComposition (VK_SPACE + VK_BACK) on trailing composed Hangul syllables (/[가-힣]$/u) in terminals before Space is pressed.

    SHA-256

    • D3RO-Voice-Setup-1.15.1-x64.exe: ce8494ecfbdc0896b5696bbfe6e2d04f1ae0f711c6516916fe280b5999dc45ab
    • D3RO-Voice-Setup-1.15.1-x64.exe.blockmap: 61669c4e202579113266c55c242b79523ee2baa615b51064f8d3b81034164d4c
    • latest.yml: 85c9e93cb0e1783d01a5f74a25f3dca90f2905a465b6bb506739cf9eba1badb2
    • latest.yml.sig: 67c53d1c0e954101f516f59de65d4696bbc6028b17eee8241ef09062c5a593d5
    Downloads