[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-77793":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":9,"totalLinesOfCode":9,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":9,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":9,"rankLanguage":9,"license":9,"archived":22,"fork":22,"defaultBranch":23,"hasWiki":22,"hasPages":22,"topics":24,"createdAt":9,"pushedAt":9,"updatedAt":37,"readmeContent":38,"aiSummary":39,"trendingCount":16,"starSnapshotCount":16,"syncStatus":40,"lastSyncTime":41,"discoverSource":42},77793,"macparakeet","moona3k\u002Fmacparakeet","moona3k","Fast, local voice app for Mac — system-wide dictation, file & YouTube transcription, and meeting recording. Powered by Parakeet TDT on Apple Silicon. Free and open-source.",null,"https:\u002F\u002Fgithub.com\u002Fmoona3k\u002Fmacparakeet","Swift",359,34,3,32,0,54,74,122,162,4.63,false,"main",[25,26,27,28,29,30,31,32,33,34,35,36],"apple-silicon","dictation","local-first","macos","neural-engine","privacy","speech-to-text","swift","transcription","meeting-recording","open-source","parakeet","2026-06-12 02:03:44","\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002FAppIcon-1024x1024.png\" width=\"128\" height=\"128\" alt=\"MacParakeet app icon\">\n\u003C\u002Fp>\n\n\u003Ch1 align=\"center\">MacParakeet\u003C\u002Fh1>\n\n\u003Cp align=\"center\">\n  Fast voice app for Mac with fully local speech and optional AI. Free and open-source.\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cem>There are many voice transcription\u002Fdictation apps, but this one is mine.\u003C\u002Fem>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fmacparakeet.com\">macparakeet.com\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdownloads.macparakeet.com\u002FMacParakeet.dmg\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FDownload-DMG-E86B3B.svg?style=for-the-badge&logo=apple&logoColor=white\" alt=\"Download DMG\">\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fdeepwiki.com\u002Fmoona3k\u002Fmacparakeet\">\u003Cimg src=\"https:\u002F\u002Fdeepwiki.com\u002Fbadge.svg\" alt=\"Ask DeepWiki\">\u003C\u002Fa>\n  \u003Ca href=\"LICENSE\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Flicense-GPL--3.0-blue.svg\" alt=\"GPL-3.0 License\">\u003C\u002Fa>\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FmacOS-14.2%2B-000000.svg\" alt=\"macOS 14.2+\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FSwift-6.0-F05138.svg\" alt=\"Swift 6\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Ftests-passing-brightgreen.svg\" alt=\"Tests passing\">\n  \u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FApple%20Silicon-only-333333.svg\" alt=\"Apple Silicon only\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002Fscreenshots\u002Ftranscribe.png?v=4\" width=\"720\" alt=\"MacParakeet — Transcribe tab with YouTube, file drop, and Record Meeting tile\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002Fscreenshots\u002Flibrary.png?v=4\" width=\"720\" alt=\"MacParakeet — Transcription library with thumbnails\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002Fscreenshots\u002Fdictations.png?v=4\" width=\"720\" alt=\"MacParakeet — Dictation Stats with streak heatmap and top apps\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002Fscreenshots\u002Fmeeting-recording.png?v=4\" width=\"720\" alt=\"MacParakeet — Live meeting recording with floating pill and Notes\u002FTranscript\u002FAsk panel\">\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Cimg src=\"Assets\u002Fscreenshots\u002Fmeeting-ask.png?v=4\" width=\"720\" alt=\"MacParakeet — Ask tab summarizing a live meeting with quick-prompt starters\">\n\u003C\u002Fp>\n\n---\n\nMacParakeet runs NVIDIA's Parakeet TDT on Apple's Neural Engine via [FluidAudio](https:\u002F\u002Fgithub.com\u002FFluidInference\u002FFluidAudio) CoreML. The v0.6 release scope includes system-wide dictation, file\u002FURL transcription, meeting recording, optional local WhisperKit recognition for languages Parakeet does not cover, and productized Transforms on `main`. All speech recognition happens on your Mac.\n\n## Release status\n\nThe [notarized DMG](https:\u002F\u002Fdownloads.macparakeet.com\u002FMacParakeet.dmg) is the stable release channel.\n\n| Channel | Status | Includes |\n|---------|--------|----------|\n| Stable DMG | Recommended for normal use | Dictation, file\u002Fvideo\u002FYouTube transcription, meeting recording, optional WhisperKit, exports, vocabulary, AI features |\n| `main` branch | Development | v0.6 release scope plus productized Transforms; hidden calendar auto-start code remains under `AppFeatures.calendarEnabled = false` |\n\nCalendar reminders, auto-start, and auto-stop are implemented in source but hidden from the v0.6 product surface while they await end-to-end validation.\n\n## What it does\n\n**Dictation** — Press a hotkey in any app, speak, text gets pasted. Hold for push-to-talk, double-tap for persistent recording. Works system-wide.\n\n**File transcription** — Drag audio or video files, or paste a YouTube URL. Full transcript with word-level timestamps, speaker labels, and export to 7 formats (TXT, Markdown, SRT, VTT, DOCX, PDF, JSON). Assign global hotkeys to trigger File or YouTube transcription from anywhere.\n\n**Meeting recording** — Record system audio and microphone together, see a live local transcript preview, take notes during the call, then save the finalized transcript to the library with export, prompts, and chat.\n\n**Text cleanup** — Filler word removal, custom word replacements, text snippets with triggers. Deterministic pipeline, no LLM needed.\n\n**AI features** — Optional summaries, chat, AI formatter, and Transforms for rewriting selected text through your configured provider. Connect any cloud provider (OpenAI, Anthropic, Gemini, OpenRouter), local runtime (Ollama, LM Studio), OpenAI-compatible endpoint, or CLI tool (Claude Code, Codex). Entirely opt-in.\n\n### Performance\n\n- ~155x realtime — 60 min of audio in ~23 seconds\n- ~2.5% word error rate (Parakeet TDT 0.6B-v3)\n- ~66 MB working memory per active Parakeet inference slot\n- 25 European languages with Parakeet auto-detection\n- Optional local WhisperKit engine for Korean, Japanese, Chinese, and many other languages\n\n### Limitations\n\n- Apple Silicon only (M1\u002FM2\u002FM3\u002FM4)\n- Parakeet is best for English and supported European languages\n- WhisperKit multilingual support requires a separate local model download before first use\n\n## Get it\n\n**Download:** Grab the [notarized DMG](https:\u002F\u002Fdownloads.macparakeet.com\u002FMacParakeet.dmg) or visit [macparakeet.com](https:\u002F\u002Fmacparakeet.com). Drag to Applications, done.\n\nFirst launch downloads the speech model (~6 GB) plus speaker-detection assets (~130 MB). Everything works fully offline after that.\n\nThe DMG is the stable release.\n\n**Standalone CLI (Homebrew):**\n\n```bash\nbrew install moona3k\u002Ftap\u002Fmacparakeet-cli\nmacparakeet-cli --version\nmacparakeet-cli health --json\n```\n\nThe Homebrew formula installs the public `macparakeet-cli` surface plus\nHomebrew-managed `ffmpeg` and `yt-dlp`. It shares the same local database and\nmodel cache as the app.\n\n**Build from source:**\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002Fmoona3k\u002Fmacparakeet.git\ncd macparakeet\nswift test\nscripts\u002Fdev\u002Frun_app.sh    # build, sign, launch\n```\n\nThe dev script creates a signed `.app` bundle so macOS grants mic and accessibility permissions. It disables target-level Xcode signing, then signs the finished bundle with the best available local identity. Override with `MACPARAKEET_CODESIGN_IDENTITY=\"Your Identity\"` if needed.\n\n**CLI:**\n\n```bash\nmacparakeet-cli transcribe \u002Fpath\u002Fto\u002Faudio.mp3\nmacparakeet-cli transcribe \u002Fpath\u002Fto\u002Faudio.mp3 --format transcript --no-history\nmacparakeet-cli models download whisper-large-v3-v20240930-turbo-632MB\nmacparakeet-cli models list\nmacparakeet-cli models select parakeet\nmacparakeet-cli transcribe \u002Fpath\u002Fto\u002Fkorean.mp3 --engine whisper --language ko --format json\nmacparakeet-cli models status\nmacparakeet-cli history\n```\n\nUse `--format transcript` for transcript-only stdout in shell pipelines. Add\n`--no-history` when you want a one-off transcription without saving a completed\nrow to MacParakeet history. `models list` and `models select` inspect or update\nthe shared speech default used by the app and `--engine app-default`. The\nWhisper CLI commands above require a downloaded local WhisperKit model. When\ndeveloping from source, prefix the same commands with `swift run`.\n\n## Tech stack\n\n| Layer | Choice |\n|-------|--------|\n| STT | Parakeet TDT 0.6B-v3 via [FluidAudio](https:\u002F\u002Fgithub.com\u002FFluidInference\u002FFluidAudio) CoreML (default) + optional local WhisperKit engine |\n| STT orchestration | Shared runtime + explicit scheduler with a reserved dictation slot and a shared meeting\u002Ffile slot; speech-engine routing and meeting-session pinning |\n| Language | Swift 6.0 + SwiftUI |\n| Database | SQLite via GRDB |\n| Auto-updates | Sparkle 2 |\n| YouTube | yt-dlp |\n| Platform | macOS 14.2+, Apple Silicon |\n\n## Vocabulary\n\nThe Vocabulary panel controls how dictated text is cleaned up before pasting. No AI involved — it's a fast, deterministic pipeline that runs in under 1ms.\n\nYou choose between two **processing modes**:\n\n- **Raw** — Paste exactly what the speech engine produces, no changes\n- **Clean** (default) — Run the text through a multi-step pipeline before pasting\n\n**The Clean pipeline** applies these steps in order:\n\n1. **Filler removal** — Strips \"um\", \"uh\", and sentence-start fillers like \"so\", \"well\", \"like\"\n2. **Custom words** — Applies your word replacement rules (e.g., \"aye pee eye\" becomes \"API\", or \"kubernetes\" gets capitalized to \"Kubernetes\"). Case-insensitive, whole-word matching. Words can be toggled on\u002Foff without deleting.\n3. **Voice Return** — If you've defined a trigger phrase (e.g., \"press return\") and speak it at the end of a dictation, it's stripped from the output and a Return keypress is simulated after paste\n4. **Snippet expansion** — Replaces short trigger phrases with longer text (e.g., \"my signature\" expands to \"Best regards, David\"). Triggers are natural language phrases because that's what the speech engine outputs. Matched longest-first to prevent collisions.\n5. **Whitespace cleanup** — Collapses spaces, fixes punctuation spacing, capitalizes the first letter\n\nEvery dictation stores both the raw and clean transcript so you can always see what changed.\n\n## AI Features\n\nAI features are entirely **opt-in** and separate from speech recognition — transcription is always local. The LLM only sees transcript text, never audio.\n\n**What it does:**\n\n- **Summarize** — After a transcription finishes, click Summarize and pick a prompt (\"Summary\", \"Action Items & Decisions\", \"Chapter Breakdown\", etc.) or write your own. The LLM processes the transcript and streams back a summary. You can generate multiple summaries per transcript, each in its own tab. Prompts marked as auto-run generate summaries automatically for new transcriptions.\n- **Chat** — Ask questions about a transcript in a multi-turn chat interface. The LLM answers based on the transcript content.\n- **AI formatter** — Optionally run your dictation and file transcripts through your AI provider to clean up grammar, punctuation, and paragraphing. Toggle on\u002Foff, customize the prompt, or reset to default.\n- **Transforms** — Select text in any app and press a bound Transform hotkey, such as `Option-1` for Polish, to rewrite the selection through your configured LLM provider.\n\n**Supported providers:**\n\n| Type | Options |\n|------|---------|\n| Cloud | Anthropic (Claude), OpenAI, Google Gemini, OpenRouter |\n| Local | Ollama, LM Studio |\n| Custom | OpenAI-Compatible (any API-shaped endpoint — vLLM, LocalAI, LiteLLM, llama.cpp server, third-party hosts) |\n| CLI subprocess | Claude Code, Codex, or another configured command |\n\n**Setup:** In Settings → AI Provider, pick a provider, enter an API key (cloud) or confirm the local server\u002FCLI command is available, select a model, and hit Test Connection. Cloud providers store keys in the macOS Keychain. Ollama and LM Studio can keep LLM inference on-device. CLI subprocess providers run the configured command locally, but that command may contact its own cloud service.\n\n## Privacy\n\nAll speech recognition runs locally. Parakeet uses the Neural Engine; the optional WhisperKit engine also runs on-device. Your audio never leaves your Mac.\n\n- **No cloud STT.** The model runs on-device. No audio is transmitted.\n- **No accounts.** No login, no email, no registration.\n- **Opt-out telemetry.** Non-identifying usage analytics and crash reporting go to a self-hosted endpoint only when telemetry is enabled. No persistent IDs, no IP storage, and no transcript\u002Faudio content is transmitted. [Source code is right here](Sources\u002FMacParakeetCore\u002FServices\u002FTelemetry\u002FTelemetryService.swift) — verify it yourself.\n- **Temp files cleaned up.** Audio deleted after transcription unless you save it.\n\n**What does use the network:** AI summaries and chat connect to configured LLM providers, or to whatever service a configured CLI tool chooses to use, when you choose them. Sparkle checks for app updates. YouTube transcription downloads video via yt-dlp. Telemetry and crash reports go to our self-hosted server unless you opt out. Core dictation and transcription stay fully offline.\n\n**Note:** Builds from source also send telemetry by default. Opt out in Settings or set `MACPARAKEET_TELEMETRY_URL` to override.\n\n## Contributing\n\n- **Report bugs** — [Open an issue](https:\u002F\u002Fgithub.com\u002Fmoona3k\u002Fmacparakeet\u002Fissues)\n- **Submit a PR** — Fork, make changes, `swift test`, open a PR\n- **Read the specs** — Architecture decisions and feature specs live in `spec\u002F`\n\nFor larger changes, open an issue first.\n\n## Support\n\nMacParakeet is free and open source. If it's useful to you, consider [sponsoring](https:\u002F\u002Fgithub.com\u002Fsponsors\u002Fmoona3k).\n\n## License\n\nGPL-3.0. Free software. [Full license](LICENSE).\n","MacParakeet 是一款针对 Mac 平台的快速本地语音应用，支持系统级听写、文件和 YouTube 视频转录以及会议录音。其核心功能包括基于苹果神经引擎运行的 NVIDIA Parakeet TDT 语音识别技术，能够实现完全本地化的语音处理，并且可选集成 WhisperKit 以支持更多语言。该应用适用于需要高效语音转文字服务、注重隐私保护的用户场景，如日常工作中的文档编辑、会议记录等。作为免费开源软件，MacParakeet 为用户提供了一个强大而安全的选择。",2,"2026-06-11 03:56:00","trending"]