[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-785":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":16,"stars7d":16,"stars30d":16,"stars90d":16,"forks30d":16,"starsTrendScore":16,"compositeScore":17,"rankGlobal":10,"rankLanguage":10,"license":18,"archived":19,"fork":19,"defaultBranch":20,"hasWiki":21,"hasPages":19,"topics":22,"createdAt":10,"pushedAt":10,"updatedAt":39,"readmeContent":40,"aiSummary":41,"trendingCount":16,"starSnapshotCount":16,"syncStatus":42,"lastSyncTime":43,"discoverSource":44},785,"openless","appergb\u002Fopenless","appergb","Hold a key, speak, release — AI-polished text appears at your cursor in any app. Open-source voice input for macOS & Windows. (按住快捷键说话，松开即得润色后的文字)","https:\u002F\u002Fgithub.com\u002Fappergb\u002Fopenless",null,"HTML",1148,93,3,21,0,18.92,"MIT License",false,"main",true,[23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38],"ai-prompt","asr","dictation","llm","macos","open-source","prompt-engineering","rust","speech-to-text","swift","tauri","typeless","typeless-alternative","voice-input","windows","wispr-flow-alternative","2026-06-12 02:00:18","\u003Cp align=\"center\">\n  \u003Cimg src=\"openless-all\u002Fapp\u002Fsrc-tauri\u002Ficons\u002F128x128@2x.png\" alt=\"OpenLess\" width=\"160\" \u002F>\n\u003C\u002Fp>\n\n\u003Ch1 align=\"center\">OpenLess\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>Open-source voice input for macOS & Windows.\u003C\u002Fstrong>\u003Cbr\u002F>\n  Press a hotkey, speak, get AI-polished text at your cursor.\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fopenless.top\">\u003Cstrong>🌐 Official site — openless.top\u003C\u002Fstrong>\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"README.md\">English\u003C\u002Fa> · \u003Ca href=\"README.zh.md\">中文\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fappergb\u002Fopenless\u002Freleases\u002Flatest\">\u003Cimg alt=\"release\" src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fv\u002Frelease\u002Fappergb\u002Fopenless?style=flat-square&color=2c5282\" \u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fappergb\u002Fopenless\u002Fblob\u002Fmain\u002FLICENSE\">\u003Cimg alt=\"license\" src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Flicense\u002Fappergb\u002Fopenless?style=flat-square&color=2f855a\" \u002F>\u003C\u002Fa>\n  \u003Cimg alt=\"macOS\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FmacOS-12%2B-1f425f?style=flat-square\" \u002F>\n  \u003Cimg alt=\"Windows\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FWindows-10%2B-0078d4?style=flat-square\" \u002F>\n  \u003Cimg alt=\"Tauri\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FTauri-2-24c8db?style=flat-square\" \u002F>\n  \u003Cimg alt=\"Rust\" src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FRust-2021-ce422b?style=flat-square\" \u002F>\n  \u003Cimg alt=\"Stars\" src=\"https:\u002F\u002Fimg.shields.io\u002Fgithub\u002Fstars\u002Fappergb\u002Fopenless?style=flat-square&color=805ad5\" \u002F>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>Join our QQ Group: 1078960553\u003C\u002Fstrong>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>Sponsors\u003C\u002Fstrong>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fwww.knin.net\" target=\"_blank\" rel=\"noopener\">\n    \u003Cimg alt=\"悠雾云数据 (Youwu Cloud Data)\" src=\"https:\u002F\u002Fwww.knin.net\u002Fupload\u002Flogo.png\" height=\"48\" \u002F>\n  \u003C\u002Fa>\n    \n  \u003Ca href=\"https:\u002F\u002Fjiangmuran.com\u002F\" target=\"_blank\" rel=\"noopener\">\n    \u003Cimg alt=\"jiangmuran\" src=\"assets\u002Fpeople\u002Fjiangmuran.png\" width=\"48\" height=\"48\" \u002F>\n  \u003C\u002Fa>\n  \u003Cbr\u002F>\n  \u003Ca href=\"https:\u002F\u002Fwww.knin.net\" target=\"_blank\" rel=\"noopener\">悠雾云数据 — www.knin.net\u003C\u002Fa>\n   · \n  \u003Ca href=\"https:\u002F\u002Fjiangmuran.com\u002F\" target=\"_blank\" rel=\"noopener\">jiangmuran — jiangmuran.com\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cstrong>Developers\u003C\u002Fstrong>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Ftripmc.top\u002F\" target=\"_blank\" rel=\"noopener\">\n    \u003Cimg alt=\"TRIP\" src=\"assets\u002Fpeople\u002Ftripmc.png\" width=\"80\" height=\"80\" \u002F>\n  \u003C\u002Fa>\n    \n  \u003Ca href=\"https:\u002F\u002Fchris233.qzz.io\" target=\"_blank\" rel=\"noopener\">\n    \u003Cimg alt=\"Chris233\" src=\"assets\u002Fpeople\u002FChris233.png\" width=\"80\" height=\"80\" \u002F>\n  \u003C\u002Fa>\n  \u003Cbr\u002F>\n  \u003Ca href=\"https:\u002F\u002Ftripmc.top\u002F\" target=\"_blank\" rel=\"noopener\">TRIP — tripmc.top\u003C\u002Fa>\n   · \n  \u003Ca href=\"https:\u002F\u002Fchris233.qzz.io\" target=\"_blank\" rel=\"noopener\">Chris233 — chris233.qzz.io\u003C\u002Fa>\n\u003C\u002Fp>\n\n---\n\nOpenLess is a cross-platform (macOS & Windows) voice-input app — a **fully open-source** alternative to commercial tools like [Typeless](https:\u002F\u002Fwww.typeless.com\u002F), [Wispr Flow](https:\u002F\u002Fwisprflow.ai), [Lazy](https:\u002F\u002Fheylazy.com), and Superwhisper. Official site: [openless.top](https:\u002F\u002Fopenless.top).\n\nPut your cursor in any text field — ChatGPT, Claude, Cursor, Notion, an email draft, a chat box — press one global hotkey and talk. OpenLess records, transcribes, polishes the text in the mode you picked, and inserts the result at the cursor. If insertion is blocked it copies to the clipboard, so the words you spoke don't get lost.\n\nUnlike voice typing tools that just dump a word-for-word transcript, OpenLess's headline mode is **AI-prompt mode**: you ramble, it adds structure, lists constraints, and produces a context-rich prompt you can paste straight into ChatGPT \u002F Claude \u002F Cursor.\n\n## A concrete example\n\nHold the hotkey, say to OpenLess:\n\n> uh… so… I want ChatGPT to write me a SQL query, from the orders table get last month's orders, group by customer, sort by amount desc, top ten\n\nRelease the hotkey. A second later your input box reads:\n\n```text\nPlease write a SQL query that:\n\n- Pulls orders from last month from the `orders` table.\n- Groups by customer.\n- Sorts by total amount, descending.\n- Returns the top 10 rows only.\n```\n\nNo edits needed. Hit Enter and ask GPT. That's the whole pitch: **write prompts with your mouth, faster and cleaner than typing them.**\n\n## Why OpenLess is open source\n\nThe closest tools are subscription SaaS: monthly bill, no bring-your-own model, your audio uploaded to the vendor, your dictionary and habits living in their account.\n\nOpenLess goes for the same end-user experience but:\n\n- **Fully open source, local-first.** Code is in this repo; all your data stays on your machine.\n- **Bring your own cloud credentials.** Volcengine streaming ASR + Ark \u002F DeepSeek-compatible chat-completions. No vendor lock-in.\n- **Tuned for AI prompts.** The \"Structured\" mode reshapes loose speech into a prompt with context, constraints, and asks — paste straight into ChatGPT, Claude, or Cursor.\n- **Won't answer for you.** The model only cleans up your text. If you say \"what features does this app still need?\", it returns that as a clean question — it does not hand you a feature list. Ask the real AI for that.\n\n## Use cases\n\n- Writing prompts for ChatGPT \u002F Claude \u002F Cursor \u002F Gemini: dictate a request, OpenLess turns it into a structured, detailed prompt.\n- Drafting emails, specs, long Slack\u002FWeChat messages: removes filler, fixes punctuation, organizes paragraphs.\n- Code comments, commit messages, PR descriptions: dump what's in your head straight to the cursor.\n- Any \"I don't want to type but I have to produce written text\" situation.\n\n## Project direction\n\nOpenLess does one thing: **turn speech into usable written text (especially AI prompts), at the current cursor.**\n\n- It does not answer questions, run tasks, or analyze your project.\n- It does not accumulate conversation context — every dictation is an independent cleanup request.\n- Speech → transcript → cleanup → insert at cursor. Clipboard fallback on failure.\n- Everything else (modes, dictionary, history, menu bar, home report) supports that one path.\n\n## Comparison\n\n| Tool | Form | How OpenLess differs |\n| --- | --- | --- |\n| [Typeless](https:\u002F\u002Fwww.typeless.com\u002F) | Closed-source macOS \u002F Windows \u002F iOS, subscription | Open source; explicit AI-prompt mode; bring-your-own ASR + LLM; data and dictionary stay on your machine |\n| [Wispr Flow](https:\u002F\u002Fwisprflow.ai) | Closed-source macOS \u002F Windows, subscription | Open source; bring-your-own ASR + LLM; transparent prompt-handling rules |\n| [Lazy](https:\u002F\u002Fheylazy.com) | Closed-source notes \u002F capture tool | Not a notes container — inserts straight into any input field |\n| [Superwhisper](https:\u002F\u002Fsuperwhisper.com) | Closed-source macOS, subscription | Open source; cloud ASR today, local ASR on the roadmap |\n\n## Status (v1.2)\n\n- Tauri 2 + Rust backend + React\u002FTS frontend. macOS 12+, Windows 10+.\n- **Toggle and push-to-talk** recording modes. `Esc` cancels at any phase, including polish\u002Finsert.\n- Volcengine streaming ASR + OpenAI Whisper-compatible batch ASR; Ark \u002F DeepSeek \u002F OpenAI-compatible chat-completions for polish.\n- 4 output modes: raw, light polish, structured (**AI prompt mode**), formal.\n- Main window: Overview \u002F History \u002F Vocab \u002F Style \u002F Settings. Persistent tray icon. Mini status capsule floating on screen.\n- **Bilingual UI** — Settings → Language switches between 简体中文 and English (auto-detects on first launch).\n- **In-app auto-update** — Settings → About → Check button; signed updater artifacts via Tauri updater plugin.\n- **Single-instance lock** — prevents two OpenLess processes from racing the same hotkey edge.\n- Dictionary entries injected as Volcengine ASR `context.hotwords` and as semantic hints during polish; hits accumulate per session.\n- Platform-native global hotkey: CGEventTap on macOS, low-level keyboard hook (`WH_KEYBOARD_LL`) on Windows.\n\n## Download & install (end users)\n\nGo to [Releases](..\u002F..\u002Freleases) and download:\n- **macOS**: `OpenLess_\u003Cversion>_aarch64.dmg` — open, drag to `\u002FApplications`\n- **Windows**: `OpenLess_\u003Cversion>_x64-setup.exe` — run the installer\n\nOn first launch, grant the permissions the app requests:\n\n**macOS:**\n1. Grant Microphone access.\n2. Grant Accessibility access.\n3. **Quit and reopen the app** — Accessibility only takes effect after a restart.\n4. Open Settings → fill in Volcengine ASR + Ark credentials.\n\n**Windows:**\n1. Grant Microphone access when prompted.\n2. Open Settings → Permissions to verify the global hotkey listener is active.\n3. Fill in Volcengine ASR + Ark credentials in Settings.\n\nFull end-user walkthrough: [USAGE.md](USAGE.md).\n\n## Build from source (developers)\n\nThe active codebase is in `openless-all\u002Fapp\u002F` (Tauri 2 + Rust + React\u002FTS). The macOS build links a vendored C ASR engine ([`antirez\u002Fqwen-asr`](https:\u002F\u002Fgithub.com\u002Fantirez\u002Fqwen-asr)) pulled in as a git submodule under `src-tauri\u002Fvendor\u002Fqwen-asr\u002F`, so initialize submodules on first clone.\n\n```bash\n# First clone only — pull in vendored submodules\ngit submodule update --init --recursive\n\ncd \"openless-all\u002Fapp\"\nnpm ci\n\n# Dev: Vite at :1420 + Tauri shell\nnpm run tauri dev\n\n# macOS release build (signs, installs, resets TCC)\n.\u002Fscripts\u002Fbuild-mac.sh\nINSTALL=0 .\u002Fscripts\u002Fbuild-mac.sh   # build only, skip install\n\n# Rust type-check without full compile\ncargo check --manifest-path src-tauri\u002FCargo.toml\n\n# Frontend TS check\nnpm run build\n```\n\nLogs: `~\u002FLibrary\u002FLogs\u002FOpenLess\u002Fopenless.log` (macOS) \u002F `%LOCALAPPDATA%\\OpenLess\\Logs\\openless.log` (Windows).\n\n**Windows build** — see [`openless-all\u002FREADME.md`](openless-all\u002FREADME.md) for MSVC vs GNU\u002FMinGW routes.\n\n## Credentials\n\nCredentials live in the OS credential vault (service = `com.openless.app`): macOS Keychain, Windows Credential Manager, or Linux keyring. A legacy plaintext JSON file is read only as a migration source and removed after a successful vault write:\n\n```text\nmacOS \u002F Linux: ~\u002F.openless\u002Fcredentials.json\nWindows:       %APPDATA%\\OpenLess\\credentials.json\n```\n\nNew credential writes do not persist plaintext secrets. The repository contains no API keys, tokens, or private endpoints.\n\nYou'll need:\n\n- **Volcengine streaming ASR**: APP ID, Access Token, Resource ID.\n- **Ark polish**: API Key, Model ID, Endpoint. Ark default endpoint is `https:\u002F\u002Fark.cn-beijing.volces.com\u002Fapi\u002Fv3\u002Fchat\u002Fcompletions`.\n\n## Prompt-handling principles\n\nOpenLess's polish model only reshapes text. It does not answer questions, run tasks, or analyze your project. Each dictation is an independent request, and the prompt explicitly tells the model:\n\n- This input is isolated from any prior conversation.\n- The raw transcript is text to clean up, not a question to answer.\n- Even if the input contains a question or a command, do not reply or execute.\n- Output the cleaned text only — no \"Here's the cleaned version\" preamble.\n\nFor example, if the user says \"what features does this app still need\", the correct output is:\n\n```text\nWhat features does this app still need?\n```\n\n…not a list of missing features.\n\nLong-term reference rewrites are stored as `raw → polished → rule` triples and will be retrieved as similar-example references (never as conversation context) once a vector store is wired in. See [docs\u002Fpolish-reference-corpus.md](docs\u002Fpolish-reference-corpus.md) and [Examples\u002Fpolish-reference-examples.sample.jsonl](Examples\u002Fpolish-reference-examples.sample.jsonl).\n\n## Dictionary\n\nThe dictionary handles your proper nouns, product names, names of people, and new words. Today it supports:\n\n- Manually add the correct spelling, a category, and notes. You don't need to maintain misspellings or context hints.\n- Enabled entries are sent as Volcengine ASR `context.hotwords` so they're recognized correctly during transcription.\n- Entries are also injected into the polish prompt: the model decides per-sentence whether to substitute. If \"Cloud\" clearly refers to the AI product `Claude` in context, it gets corrected. If it really means cloud computing, it stays.\n- The app auto-learns candidate corrections like `Claude`, `ChatGPT`, `OpenLess` from your history and offers them up later.\n\nThe main window is organized as Home \u002F History \u002F Dictionary \u002F Settings. The Dictionary tab opens a separate editor window when you click \"New\". The Home tab shows total dictation time, total characters, average chars-per-minute, estimated time saved, and dictionary participation stats.\n\n## Architecture\n\nThe active implementation is Tauri 2 (`openless-all\u002Fapp\u002F`). Auto-updates ride on the Tauri updater plugin; signed updater artifacts are produced by CI on every `v*-tauri` tag.\n\n**Tauri backend (Rust)** — each module depends only on `types.rs`:\n\n```\ntypes.rs         Pure value types: DictationSession, PolishMode, HotkeyBinding, errors\nhotkey.rs        Global hotkey (CGEventTap on macOS, WH_KEYBOARD_LL on Windows, rdev on Linux)\nrecorder.rs      Mic → 16 kHz mono Int16 PCM, RMS callback\nasr\u002F             Volcengine streaming ASR (WebSocket) + Whisper HTTP\npolish.rs        OpenAI-compatible chat-completions (Ark \u002F DeepSeek \u002F etc.)\ninsertion.rs     AX focused-element → clipboard + Cmd+V → copy-only fallback\npersistence.rs   History \u002F preferences \u002F vocab JSON + platform credential vault\npermissions.rs   TCC checks (Accessibility \u002F Microphone)\ncoordinator.rs   State machine: Idle → Starting → Listening → Processing\ncommands.rs      Tauri IPC surface\n```\n\n**React frontend (`src\u002F`)** — state via Recoil atoms (`pages\u002F_atoms.tsx`); hotkey capability\u002Fbinding via `HotkeySettingsContext`; all backend calls go through `lib\u002Fipc.ts`.\n\nThe dictation pipeline: `hotkey edge → Recorder.start + ASR.openSession → [audio frames] → hotkey edge → Recorder.stop + ASR.sendLastFrame → Polish → Insert → History.save`.\n\nSee [CLAUDE.md](CLAUDE.md) for invariants and module-wiring rules.\n\n## Roadmap\n\nPlanned but not yet shipped:\n\n- Dictation translation mode: hold a separate hotkey, speak in your language, insert in target language ([#43](..\u002F..\u002Fissues\u002F43)).\n- Cross-session style memory: polish learns user's tone over time ([#46](..\u002F..\u002Fissues\u002F46)).\n- Snippets (no UI \u002F trigger logic yet).\n- History enhancements: copy button, search, re-polish, re-insert.\n- \"Paste last result\" hotkey.\n- Multi-monitor capsule placement on the focused screen.\n\n## Maintainer release checklist\n\n- Bump version in `openless-all\u002Fapp\u002Fpackage.json`, `src-tauri\u002Ftauri.conf.json`, and `src-tauri\u002FCargo.toml`.\n- Run `INSTALL=0 .\u002Fscripts\u002Fbuild-mac.sh` and confirm the `.app` launches.\n- Verify on a clean macOS box: permission flow, hotkey, recording, ASR, polish, insertion, clipboard fallback.\n- Push a `v\u003Cversion>-tauri` tag — CI builds + signs the updater artifacts and the macOS `.dmg` + Windows `.msi`. The updater needs `TAURI_SIGNING_PRIVATE_KEY` repo secret (matching the pubkey in `tauri.conf.json`).\n\n## Acknowledgements\n\nOpenLess sincerely thanks our sponsors, developers\u002Fcontributors, and the broader LinuxDo community.\n\nWe appreciate sponsors for making sustained project work possible, and we thank developers and contributors for building, reviewing, and improving OpenLess.\n\nOpenLess also recognizes and appreciates the LinuxDo community for its open, practical, and developer-friendly atmosphere. Many ideas, discussions, and early feedback around OpenLess were inspired by the broader open-source spirit represented by LinuxDo.\n\nThis acknowledgement does not imply official endorsement or affiliation.\n\n## License\n\nMIT\n","OpenLess 是一个适用于 macOS 和 Windows 的开源语音输入工具，用户只需按住快捷键说话，松开后即可在光标处获得经过 AI 优化的文字。其核心功能包括实时语音转文字、AI 文本润色，并且支持自定义快捷键。项目采用 Tauri 框架开发，结合了 Rust 和 Swift 等现代编程语言以确保高效性能与跨平台兼容性。特别适合需要频繁进行文本录入但又希望减少键盘输入负担的场景，比如写作、笔记整理或是远程会议记录等。",2,"2026-05-06 17:27:50","CREATED_QUERY"]