Skip to content

release: 0.5.2 #8

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
RGB-loop merged 2 commits into from
Oct 17, 2026
Merged

release: 0.5.2 #8

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
e9203cf
release: 0.5.2 — local API + MCP, deep links, voice input (macOS), OC…
RGB-loop Oct 17, 2026
345970d
build: keep local worktrees (.claude/), .git, server/ deps and test o…
RGB-loop Oct 17, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
11 changes: 11 additions & 0 deletions CHANGELOG.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,8 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0

## [Unreleased]

## [0.5.2] - 2026-10-18

### Added
- **Deep links for your notes (`vixel://open?…`).** Right-click any item (or use the
inspector) → Copy Link / Copy Markdown Link, paste into Obsidian / Notes / anywhere;
Expand Down Expand Up @@ -36,6 +38,15 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
points you to System Settings → Keyboard → Dictation instead of downloading anything itself.
Windows is not covered yet (no offline engine met the quality bar).

### Fixed
- **Text in images: long lines lost most of their characters** (#3). Each recognized line
was squeezed into a 320 px wide input, so a 30–40 character line (screenshots, articles)
came back with large gaps. Lines now get a width that matches their aspect ratio, as in
PaddleOCR itself; the reporter's screenshot now reads line for line like their local
PaddleOCR. Existing libraries re-run OCR automatically after updating.
- The player's segment strip said "片段 2" for a two-segment video, which read like
"segment #2"; it now says "2 个片段".

### Decided against (for now)
- **Search by a sound clip.** Measured on the demo footage: audio queries against the joint
frames+audio segment vectors rank the right video first only 3/12 times (the picture
Expand Down
2 changes: 2 additions & 0 deletions electron-builder.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -20,6 +20,8 @@ files:
# 构建产物 / 模型 / 原生运行时都不进 asar：模型和 LiteRT 走 extraResources，dist 是 CLI，release 是上一次的打包输出
# 注意：不要在 mac: / win: 下写 files —— 平台级 files 会整个替换这里的列表，上面的排除全部失效
- '!{release,dist,resources}/**'
# 本地工作目录：worktree（含整份源码和资源副本）、server 的 Node ABI 依赖、自定义网页构建、测试输出
- '!{.claude,.git,server,e2e-out,coverage}/**'
# onnxruntime-node（人脸 / OCR）自带全部平台的二进制，只留目标平台的
- '!node_modules/onnxruntime-node/bin/napi-v*/!(${platform})/**'
asar: true
Expand Down
4 changes: 2 additions & 2 deletions package-lock.json
View file
Open in desktop

Some generated files are not rendered by default. Learn more about how customized files appear on GitHub.

2 changes: 1 addition & 1 deletion package.json
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,6 +1,6 @@
{
"name": "vixel",
"version": "0.5.1",
"version": "0.5.2",
"description": "Private, offline AI search for your photos, videos and audio",
"main": "./out/main/index.js",
"author": "Vixel Team",
Expand Down
Loading