VocalCode 1.4.1

VocalCode

Free, local-first dictation and meeting notes, open source under
AGPL-3.0-only. There is nothing to buy: no purchase, account or activation
code. Recognition runs on your CPU after the selected model downloads.

New in 1.4.1

• One app, free for good. The early free builds (VocalCode Community 1.3.1
  and 1.4.0) and the paid releases are now one app, just VocalCode. A paid
  installation updates to it in place and keeps everything (see below).
• Setup in three steps: pick your language (its model starts
  downloading), check your talk key (add another if your keyboard has no
  Right Ctrl), then try a sentence in a box that shows what VocalCode heard.
• New installs start with safer defaults: the mouse Back button is no
  longer Enter (tick it in setup or Shortcuts if you want it); the speech
  gate is on, so a noise-only press types nothing; History is kept on disk,
  encrypted, for a week (change it on the History page). Existing installations keep
  their current settings.
• Notices while you dictate. Pressing the talk key while the model is
  still downloading, the microphone is broken or no text field has focus now
  shows a short line at the indicator instead of doing nothing.
• Model downloads resume after a dropped connection or quitting, instead
  of starting over, and a failed download says why and retries.
• English defaults to Parakeet on computers fast enough for it (better
  with coding words than SenseVoice); smaller computers keep SenseVoice.
• Faster in noisy rooms: pauses are found against the room's own noise
  floor, so long dictations are prepared while you talk rather than all after
  you let go.
• Chinese punctuation in mixed Chinese/English sentences stays
  full-width; a result that is only punctuation (a stray "。" over noise) is
  treated as nothing heard.
• Qwen3-ASR: dictations longer than about 38 seconds are no longer cut
  off or replaced by the word "language"; long input is decoded in pieces,
  and text invented over silence or hum is dropped.
• About & help replaces the old licence page, with Copy diagnostics
  (nothing is sent). If VocalCode closes unexpectedly, the next start says so
  and points to the log.
• Tested with speech: the release voice corpus now has 1,199 clips,
  including mixed Chinese/English, noise-only presses and 30–90 second
  dictations, replayed through the production pipeline on four model routes.

Downloads

• Windows x64: VocalCodeSetup.exe (publisher: Daming Wu).
• Apple-silicon macOS: VocalCode-<version>.dmg (signed and notarized).
• Exact Corresponding Source: VocalCode-source-<version>.tar.gz.
• SHA256SUMS: hashes of release assets; latest.json: the app's update feed.

Installing over an earlier VocalCode

VocalCode is now a single free app. It installs where the paid VocalCode
releases (1.2.1 and earlier) were installed and uses the same data folder, so
it replaces such an installation in place: settings, dictionary, snippets,
meetings and downloaded models stay where they are, and nothing needs
importing. A paid release whose licence or trial is still active offers this
version as an in-app update; otherwise run the installer over it (on a Mac,
replace VocalCode.app in Applications). The paid licence and trial files are
left where they are, and VocalCode never reads them. Uninstalling VocalCode
later keeps your data folder.

A paid VocalCode installed with Scoop cannot update itself this way: run
scoop uninstall vocalcode, then install from this page.

The early free builds, VocalCode Community 1.3.1 and 1.4.0, were a separate
app. On Windows, the installer uninstalls VocalCode Community 1.3.1 or 1.4.0
and keeps its data folder; Settings → System → Previous VocalCode copies
what you choose from it. On a Mac, remove VocalCode Community 1.3.1 or 1.4.0
yourself. Those builds cannot update to this version in the app: download it
from this page once. Don't run two copies at once: each would type every
dictation.

Checks

Before publication, every release is tested for installation, upgrade,
signatures, notarization, native linkage and data preservation, and on
Windows for an in-place update of a real paid VocalCode 1.2.1 through its own
updater. Actual microphone, system-audio, meeting and accessibility behavior
still depends on the device and operating-system permissions; these checks do
not promise perfect speech recognition or replace broad native-device testing.

VocalCode updates from this project's GitHub Releases and verifies the exact
size, SHA-256, publisher signature and product identity before installing.

中文

1.4.1 新增：早期免费版（VocalCode Community 1.3.1 和 1.4.0）与付费版合并为同一个应用 VocalCode，付费版可原地更新、数据不动；
首次设置改成三步（选语言、确认说话键、试说一句）；新安装默认：鼠标后退键不再当回车、语音闸门开启
（只有噪声时不输入）、历史记录加密保存在本机一周，已有安装保持原设置；模型没准备好、麦克风异常或没有
输入框时，指示器会给出提示；模型下载可断点续传；性能足够的电脑上英语默认用 Parakeet；嘈杂环境里更早
开始识别；中英混说时中文标点保持全角，噪声里只识别出一个标点时不再输入；Qwen3-ASR 超过约 38 秒的
长听写不再被截断或变成 "language"；新增「关于与帮助」和意外退出提示。发布前用 1199 段合成语音
（含中英混说、纯噪声、30–90 秒长听写）在四条识别路线上做了完整回放测试。

VocalCode 现在是一个免费应用，按 AGPL-3.0 开源，无需购买、账户或激活。下载：Windows 用
VocalCodeSetup.exe，Apple 芯片 Mac 用 VocalCode-<版本>.dmg，源码为
VocalCode-source-<版本>.tar.gz。两个平台的安装包均经过签名，Mac 包完成 Apple 公证。

VocalCode 安装在付费版（1.2.1 及更早）原来的位置，使用同一个数据文件夹，所以会原地替换付费版：
设置、词典、片段、会议记录和已下载的模型原样保留，无需导入；付费版的授权与试用文件保持原样，
VocalCode 从不读取。授权或试用仍有效的付费版会在应用内提示这次更新，其他情况请直接覆盖安装
（Mac 上替换“应用程序”里的 VocalCode.app）。之后卸载 VocalCode 也会保留数据文件夹。
通过 Scoop 安装的付费版无法这样更新：请先运行 scoop uninstall vocalcode，再从本页安装。

早期免费版 VocalCode Community 1.3.1 和 1.4.0 是单独的应用。在 Windows 上，安装程序会卸载
VocalCode Community 1.3.1 或 1.4.0 并保留它的数据文件夹，可在 设置 → 系统 → 旧版 VocalCode
里选择要复制的内容；在 Mac 上请自行删除 VocalCode Community 1.3.1 或 1.4.0。这两个早期版本无法在
应用内更新到本版本，请从本页下载一次。不要同时运行两个副本，否则每次听写都会被输入两遍。
首次下载模型和检查更新需要联网；录音与识别在本地处理。

Maintainer: Daming Wu.
Issues: https://github.com/wudaming00/vocalcode-community/issues

VocalCode Community 1.4.0

VocalCode Community

Free, local-first dictation and meeting notes under AGPL-3.0-only. No purchase,
account or activation code is required. Recognition runs on your CPU after
the selected model downloads.

New in 1.4.0

• Writing page. Opt-in voice commands for complete dictations: "new
  line" / "new paragraph" (换行 / 另起一段), "scratch that" (删掉上一句),
  spoken lists ("first… second…", 第一，第二…), coding words ("camel case user
  id" → userId, "open paren"), and ending with "press enter" (回车) to send.
  Formal / Casual / Very casual style, per program, with a chat-app preset and
  a Try-it box that runs the same rules as dictation. Everything is off until
  you turn it on; History keeps the original recognition.
• Double-tap to lock. Hold to talk as before, or double-tap the talk key to
  keep listening hands-free; press once more to finish.
• Mute other audio while dictating (Windows, opt-in), restoring exactly what
  it muted; skipped during meeting capture.
• Home insights: day streak, words per minute, words this week and a
  12-week heatmap, from counts only.
• Rewrite scratchpad: translate to English or Chinese, or give your own
  instruction (you can dictate it). You still review every candidate.
• Desktop control capsule (Windows, opt-in) for starting, stopping and
  cancelling dictation with the mouse.
• Fixes: multi-line text and snippets are pasted rather than typed, so a
  line break no longer presses Enter and sends half a chat message; a
  Qwen3-ASR result containing a line break or a leaked
  "language English<asr_text>" header no longer drops or garbles the
  dictation; Spanish, French and German cover every string.
• Tested with speech: a new voice-corpus release test replays 1,072
  synthetic clips (generic neural voices, accents, noise, speed and low
  volume) through the production pipeline on four speech-model routes before
  release. Spoken coding words ("snake case", "open paren") are recognized
  best by the Qwen3-ASR and Parakeet English models; the default SenseVoice
  model often mishears them.

Downloads

• Windows x64: VocalCodeCommunitySetup.exe (publisher: Daming Wu).
• Apple-silicon macOS: VocalCodeCommunity-<version>.dmg (signed and notarized).
• Exact Corresponding Source: the attached versioned source archive.
• SHA256SUMS: hashes of release assets; latest.json: the app's update feed.

Community installs separately from the previous paid edition, with its own
data folder, startup entry and update channel. Existing data is neither moved
nor deleted. Close the other edition before dictating to avoid two global
hotkey listeners. Dictionary/snippets can be exported and imported explicitly.

The first community installer has automated installation, upgrade, signature,
notarization, native-linkage and data-preservation checks. Actual microphone,
system-audio, meeting and accessibility behavior still depends on the device
and operating-system permissions; these checks do not promise perfect speech
recognition or replace broad native-device testing.

Community updates use GitHub Releases and still verify exact sizes/hashes and
the expected platform publisher and product identity before installation.
The previous paid updater does not silently migrate users to this edition.

中文

1.4.0 新增：「写作」页（可选的语音换行、删掉上一句、口述列表、代码词、结尾说「回车」发送，
正式／随意／很随意三档风格可按程序设置，并带「试一试」）；双击说话键锁定免提；听写时静音
其他声音（Windows，可选）；首页统计（连续天数、每分钟字数、12 周热力图，只记数量）；
改写草稿支持翻译和自定义指令；桌面控制胶囊（Windows，可选）。多行文字和片段改为粘贴，
不再把换行当回车发出半句话；Qwen3-ASR 输出里带换行或漏出提示头时，不再丢掉整句或打出乱码。
发布前用 1072 段合成语音在四条识别路线上做了完整回放测试。

社区版免费使用全部本地功能，无需激活。Windows 与 Apple 芯片 Mac 安装包
均经过签名；Mac 包完成 Apple 公证。社区版独立安装，不覆盖旧版或迁移／删除旧数据。
开始听写前请退出另一个版本，避免快捷键监听冲突。词典与片段可由用户主动导入。
首次下载模型和检查更新需要联网；录音与识别在本地处理。

Maintainer: Daming Wu.
Issues: https://github.com/wudaming00/vocalcode-community/issues

VocalCode Community 1.3.1

VocalCode Community

Free, local-first dictation and meeting notes under AGPL-3.0-only. No purchase,
account or activation code is required. Recognition runs on your CPU after
the selected model downloads.

Downloads

• Windows x64: VocalCodeCommunitySetup.exe (publisher: Daming Wu).
• Apple-silicon macOS: VocalCodeCommunity-<version>.dmg (signed and notarized).
• Exact Corresponding Source: the attached versioned source archive.
• SHA256SUMS: hashes of release assets; latest.json: the app's update feed.

Community installs separately from the previous paid edition, with its own
data folder, startup entry and update channel. Existing data is neither moved
nor deleted. Close the other edition before dictating to avoid two global
hotkey listeners. Dictionary/snippets can be exported and imported explicitly.

The first community installer has automated installation, upgrade, signature,
notarization, native-linkage and data-preservation checks. Actual microphone,
system-audio, meeting and accessibility behavior still depends on the device
and operating-system permissions; these checks do not promise perfect speech
recognition or replace broad native-device testing.

Community updates use GitHub Releases and still verify exact sizes/hashes and
the expected platform publisher and product identity before installation.
The previous paid updater does not silently migrate users to this edition.

中文

社区版免费使用全部本地功能，无需激活。Windows 与 Apple 芯片 Mac 安装包
均经过签名；Mac 包完成 Apple 公证。社区版独立安装，不覆盖旧版或迁移／删除旧数据。
开始听写前请退出另一个版本，避免快捷键监听冲突。词典与片段可由用户主动导入。
首次下载模型和检查更新需要联网；录音与识别在本地处理。

Maintainer: Daming Wu.
Issues: https://github.com/wudaming00/vocalcode-community/issues

VocalCode Community — Source Preview 2026-09-22

VocalCode Community — early source preview

VocalCode's desktop source is now available under AGPL-3.0-only.
This is a source preview for developers and contributors, not a signed
installer or a production-ready replacement for an existing installation.

Included

• CPU-first local dictation, personal vocabulary/correction learning, and
  local meeting recording, transcription, search, and export source.
• Community builds enable local features without purchase or activation.
• English/Chinese project documentation, maintainer and contribution guidance,
  third-party notices, and private vulnerability reporting.
• GitHub-hosted Windows and Apple-silicon macOS tests/builds, UI regressions,
  plus a Linux dependency-audit and secret-scan job. No self-hosted runner or
  production signing credential is required for these checks.
• Rustls updated to 0.23.45; local audit reported no known vulnerabilities on
  2026-09-22. Two dependency warnings remain documented in SECURITY.md.

Before running a build

Read BUILDING.md
and release readiness.
The preview still shares the existing app's data/instance/autostart identity:
use a separate OS user for isolated testing; do not overwrite your installation.
Meeting audio, permissions, migration, signed installers, and update channels
still need native-device/release validation. Some UI copy retains legacy Pro labels.

Model weights are not included in these source archives and retain their own
licences. AGPL permits compliant commercial use; it does not require users to
publish their ordinary dictation or meeting recordings.

The published commit 19e70b47fda030c7b18901a55fb7e0d10630d820 passed
GitHub-hosted CI:
Windows and Apple-silicon macOS tests, strict Clippy, UI/documentation checks,
optimized builds and artifact uploads, plus secret/dependency checks.
This is automated build validation, not microphone/system-audio or installer testing.

CI developer artifacts, when available, are unsigned and retained briefly;
they are not supported installers. See the live
CI results
instead of treating this source announcement as a claim that every platform
has completed validation.

Maintainer: Daming Wu.
Issues ·
Contributing ·
Private security reports.

---

这是 AGPL-3.0-only 的早期源码预览，不是可直接覆盖安装的正式版本。
社区构建不需要激活；请使用独立的操作系统用户测试，不要覆盖现有数据。
本次不改变已安装版本、既有付款服务或付费购买条款。