{"slug": "show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos", "title": "Show HN: EnviousWispr, an open-source Wispr Flow alternative for macOS", "summary": "EnviousWispr, a free open-source AI dictation app for macOS, launches as a privacy-focused alternative to cloud-based services. The app runs entirely on-device using Apple Silicon, offering sub-second transcription with optional AI polish, and requires no account or subscription.", "body_md": "**Talk naturally. Paste perfectly.**\n\nFree, on-device AI dictation and speech-to-text for macOS.\n\nPowered by Apple Silicon. No cloud, no account, your voice never leaves your Mac.\n\n## ew-demo-tmp.mp4\n\nEnviousWispr is a free AI dictation app for macOS that runs entirely on-device. It uses Whisper and Parakeet speech-to-text models on Apple Silicon to transcribe your voice locally, polishes the output with an optional LLM, and pastes clean text into whatever app you're working in. Transcription is sub-second; with optional AI polish, the full hotkey-to-paste flow typically lands in around a second and a half.\n\nNo cloud. No account required. No subscription. No audio ever leaves your Mac. Works fully offline.\n\nIt is open source under the GPLv3, actively maintained, and built to be a tool you can rely on every day.\n\n| EnviousWispr | Cloud dictation services | |\n|---|---|---|\nPrivacy |\n100% on-device transcription | Audio uploaded to servers |\nSpeed |\nSub-second transcription, paste-on-stop | Network round-trip latency |\nModels |\nParakeet v3 (NVIDIA NeMo) + WhisperKit (OpenAI Whisper) | Single vendor model |\nPolish |\nOptional. Fully on-device (EG-1, Apple Intelligence, Ollama) or bring-your-own-key cloud (GPT, Gemini) | Cloud polish, included in subscription |\nCost |\nFree. No account, no subscription | Monthly subscription |\nWorks offline |\nYes, fully functional without internet | No |\n\n``` php\nPress hotkey  -->  Record  -->  Transcribe  -->  Polish (optional)  -->  Paste\n    ~0ms          live        ~400-800ms         ~200-500ms            instant\n```\n\n**Press your hotkey** from any app. Push-to-talk, toggle, or hands-free (double-press to lock for long-form), your choice.**Speak naturally.** Silero VAD detects when you stop talking and ends recording automatically.**On-device transcription.** Choose Parakeet v3 (fastest, 25 European languages) or WhisperKit (99 languages, with automatic language detection).**AI polish**(optional). Clean up grammar, punctuation, and formatting. Runs fully on-device with EG-1 (our own custom model), Apple Intelligence (macOS 26+), or Ollama, or in the cloud via OpenAI or Gemini with your own API key.**Text lands in your clipboard** and optionally auto-pastes into the active app.\n\nSee the full interactive pipeline demo at\n\n[enviouswispr.com/how-it-works]\n\nDictation is only useful if you can trust it mid-sentence, every time. EnviousWispr keeps the critical path (record, transcribe, paste) deliberately separate from every optional enhancement, so a hiccup in a \"nice to have\" can never swallow your words. When an optional step cannot run, you simply get your raw transcribed text instead of an error.\n\n| What we hardened | What it means for you |\n|---|---|\nDelivery survives non-critical failures |\nIf saving to your history cannot complete (full disk, permissions), your dictation is still pasted. The save is best-effort and never blocks delivery. |\nPaste that actually lands |\nA multi-step delivery path tries the fastest reliable method first and falls back automatically, so text lands even in apps that resist the usual paste (Word, Excel, Pages, Numbers, and more). |\nOnboarding that won't leave you half-set-up |\nSetup won't let you start until Accessibility is granted, and it re-checks if you later revoke permission. |\nClear answers when AI polish has a problem |\nIf a cloud or local model fails (OpenAI, Gemini, Ollama), you get a specific, plain-language message, and your raw text still arrives. |\nDeterministic cleanup before AI |\nFor English, numbers, dates, and money are formatted by a fixed, predictable step, even when AI polish is off or unavailable. |\nFast recovery after idle |\nAfter the app sits idle, it re-wakes in a fraction of a second so your next press, and its first word, are not lost. |\nPrivacy-safe diagnostics |\nCrash reports carry counts and context, never your transcript or audio, and are redacted before they are sent. |\nHardened releases |\nEvery build is signed, notarized, and Gatekeeper-checked before it ships. |\n\n| Model | Best for | Languages | Disk space | Runs on |\n|---|---|---|---|---|\nParakeet TDT v3 |\nFastest dictation (default) | 25 European languages | ~460 MB | Apple Neural Engine |\nWhisperKit (Whisper Large v3 Turbo) |\nBroadest language coverage and automatic language detection | 99 languages | ~1.6 GB | Apple GPU |\n\nBoth models run entirely on-device on Apple Silicon using CoreML. Parakeet runs on the Apple Neural Engine, which is what makes the default engine near-instant; WhisperKit runs on the GPU for broad-language accuracy. First launch downloads and compiles the model; subsequent launches are instant.\n\nTranscription gets your words down. AI polish cleans them up: it drops filler, fixes grammar and punctuation, and structures rambling speech into readable text. This step is optional, and by default it never leaves your Mac.\n\n| Polish engine | What it is | Runs on | Extra download |\n|---|---|---|---|\nEG-1 (recommended) |\nOur own model, custom fine-tuned for dictation cleanup | On-device, macOS 14+ | ~2.9 GB (optional) |\nApple Intelligence |\nApple's on-device model, no extra download | On-device, macOS 26+ | none |\nOllama |\nBring your own local model (3B or larger recommended) | On-device | varies |\nOpenAI / Gemini |\nBring-your-own-key cloud polish, text only | Cloud (your key) | none |\n\n**EG-1** is our own AI model, fine-tuned specifically for dictation cleanup and optimized for Apple Silicon. It runs entirely on your Mac with no internet required, and it closes the gaps a general on-device model leaves: reliably turning a spoken list into a real list, splitting a wall of speech into clean paragraphs, and keeping only the corrected version when you fix yourself mid-sentence. Because it is our own model rather than Apple's, it works across the full supported range (macOS 14 and later), not just macOS 26. EG-1 is distributed under its own model license, not the GPLv3 that covers the app code (see [License](#license)).\n\nOn our own benchmark of 1,890 real dictation-cleanup cases, EG-1 passed 93.7%, ahead of both GPT-5.4-mini (83.8%) and Gemini 3.5 Flash (92.6%) on the same cases with the same judge. This is our own benchmark, not an independent review. The eval harness and the exact prompts are public in [ scripts/eval/](/saurabhav88/EnviousWispr/blob/main/scripts/eval) so you can inspect or rerun them; the test cases are personal dictations and stay private.\n\n- 🎙️\n**Dual ASR engines** with[Parakeet v3](https://github.com/FluidInference/FluidAudio)(NVIDIA NeMo) and[WhisperKit](https://github.com/argmaxinc/argmax-oss-swift)(OpenAI Whisper) - ✨\n**AI polish that respects your words**: strips filler words and false starts, fixes grammar and punctuation, formats numbers, dates, and URLs, and honors your custom vocabulary, all in your spoken language (never translated or rewritten) - 🔒\n**Polish that can stay private**: run it fully on-device with EG-1 (our own custom model), Apple Intelligence (macOS 26+), or Ollama, or in the cloud via OpenAI GPT / Google Gemini with your own API key - 🌍\n**Multilingual with automatic language detection**: speak in any supported language and EnviousWispr detects it, then offers to lock it in for faster, more accurate transcription - 😀\n**Speak an emoji**: say the emoji's name followed by \"emoji\" (like \"thumbs up emoji\") and the glyph drops right in - ✋\n**Voice Activity Detection** via Silero VAD that stops recording automatically when you stop talking - 📚\n**Custom vocabulary and vocabulary packs** for names, brands, and technical terms the ASR might miss, plus one-tap import of names from your Contacts (which never leave your Mac) - ⌨️\n**Global hotkey** with push-to-talk, toggle, and hands-free modes (double-press to lock for long-form dictation) - 📋\n**Auto-paste** directly into the active app, or just copy to clipboard - 🕘\n**Transcript history** for browsing, searching, and reviewing past dictations - 🧭\n**Menu bar native** with minimal footprint - 🔄\n**Auto-updates** via Sparkle\n\nEnviousWispr ships often. A few of the user-facing improvements from recent releases:\n\n**Pasting lands in more apps.** Text now reliably reaches apps that previously said \"Copied\" but never pasted (Word, Excel, Pages, Numbers, OneNote, and others). (v2.1.4)**Vocabulary packs and Contacts import.** Turn on a pack for brands and jargon, or import the names of people you know in one tap, so hard-to-spell names come out right. Your contacts never leave your Mac. (v2.1.3)**No swallowed first word after a break.** After idling, the engine re-wakes in a fraction of a second and captures your words, including the very first one. (v2.1.2 and v2.1.3)**Soft and distant speech captured.** Quiet, whispered, or far-from-the-mic speech is now transcribed instead of dropped. (v2.1.2)**Automatic update checks.** The app looks for new versions on its own, with a clear \"Check for Updates\" control in Settings. (v2.1.2)**Clearer AI polish errors.** Cloud and local polish failures (OpenAI, Gemini, Ollama) now show a specific, actionable message, and your raw text still arrives.\n\nSee the [full release history](https://github.com/saurabhav88/EnviousWispr/releases) for every version.\n\nInstall with [Homebrew](https://brew.sh):\n\n```\nbrew install --cask saurabhav88/tap/enviouswispr\n```\n\nOr download manually:\n\n- Download\n[EnviousWispr.dmg](https://enviouswispr.com/download?source=github_readme&utm_source=github&utm_medium=referral&utm_campaign=enviouswispr-evergreen-readme)from the latest release - Drag to Applications, launch\n- Grant\n**Microphone**,** Accessibility**, and (on first paste fallback)** Automation**permissions when prompted - Set your preferred hotkey in Settings > Shortcuts\n- Start talking\n\n**Optional:** Turn on AI polish in Settings > AI Polish. Keep it fully on-device with Apple Intelligence (macOS 26+) or Ollama, or add an OpenAI or Gemini API key.\n\n- macOS 14 (Sonoma) or later\n- Apple Silicon (M1 or newer)\n\nCore dictation works across the full supported range. The built-in Apple Intelligence polish option requires macOS 26 or later; on earlier versions dictation works normally and you can use Ollama or a cloud key for polish instead.\n\n```\ngit clone https://github.com/saurabhav88/EnviousWispr.git\ncd EnviousWispr\nswift build            # compiles the Swift packages (dependencies resolve via SPM)\n```\n\nThe runnable `.app`\n\nis assembled by the Xcode build engine via Tuist, not by `swift build`\n\n. Use `./scripts/build-dev-app.sh`\n\nfor a local dev build, or the release path below. First build takes several minutes as ML models compile.\n\nFor a distributable `.app`\n\nbundle and DMG:\n\n```\n./scripts/build-release-dmg.sh <version>\n```\n\nThe release build runs on the Xcode engine via Tuist, so it requires full Xcode (26+) plus mise and Tuist; set `CODESIGN_IDENTITY`\n\nto sign. Running the app itself requires macOS 14+.\n\nThe app follows a pipeline state machine: **idle --> recording --> transcribing --> polishing --> complete**.\n\nKey design choices:\n\n**Swift 6 strict concurrency** with full actor isolation**Dual pipeline architecture** with deliberately separate Parakeet and WhisperKit backends (isolation is a feature, not tech debt)**Heart & Limbs pattern** where the critical path (audio, ASR, paste) never fails, and features (polish, custom words, filler removal) degrade gracefully**Local-first** with LLM polish as an opt-in enhancement using your own keys\n\nContributions are welcome. EnviousWispr is open source under the GPLv3. Please open an issue to discuss significant changes before submitting a PR.\n\nThis project uses conventional commits: `feat(scope):`\n\n, `fix(scope):`\n\n, `refactor(scope):`\n\n.\n\nEnviousWispr is built on a simple principle: **your voice is yours.**\n\n- Audio is captured, transcribed, and discarded locally. Nothing is uploaded, stored, or shared.\n- LLM polish (if enabled) can run entirely on your Mac with EG-1 (our own model), Apple Intelligence, or a local Ollama model, so the polish step makes no network call. If you pick a cloud provider (OpenAI or Gemini), only text is sent (your transcript plus the polish instructions) using your own API key. Audio is never sent.\n- Anonymous product analytics (PostHog) can be disabled in Settings.\n- Crash reporting (Sentry) contains no transcript content, audio, or personal data.\n\n**Website:**[enviouswispr.com](https://enviouswispr.com)** X:**[@EnviousLabs](https://x.com/EnviousLabs)** Email:**[hello@enviouswispr.com](mailto:hello@enviouswispr.com)\n\nBuilt by [Envious Labs](https://enviouslabs.co)\n\nEnviousWispr is open source under the [GNU General Public License v3](/saurabhav88/EnviousWispr/blob/main/LICENSE) (GPLv3), an OSI-approved license. You can read, build, modify, and redistribute the code under the terms of the GPL, including for commercial purposes; distributed derivative works must also be licensed under the GPLv3 with their source available.\n\nCopyright (C) 2024-2026 Envious Labs LLC.\n\n**The EG-1 model is not open source.** The GPLv3 covers the EnviousWispr application code only. EG-1's model weights are not part of this repository; they download separately and are distributed under the [EG-1 Community Model License](/saurabhav88/EnviousWispr/blob/main/EG-1-MODEL-LICENSE.txt). You are free to download and use EG-1 within EnviousWispr, but you may not redistribute, re-host, mirror, or bundle the model into other products, or use the model, or outputs generated by it at scale, to train, fine-tune, or distill another model intended for commercial distribution. EG-1 is a fine-tuned derivative of Qwen3-4B-Instruct-2507 (Apache-2.0); this license applies to the fine-tuned weights.\n\nThe EnviousWispr name and logo are trademarks of Envious Labs and are not covered by the GPL.", "url": "https://wpnews.pro/news/show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos", "canonical_source": "https://github.com/saurabhav88/EnviousWispr", "published_at": "2026-07-19 00:53:00+00:00", "updated_at": "2026-07-19 01:21:14.163422+00:00", "lang": "en", "topics": ["artificial-intelligence", "developer-tools"], "entities": ["EnviousWispr", "Apple", "Whisper", "Parakeet", "NVIDIA NeMo", "OpenAI", "Gemini", "Ollama"], "alternates": {"html": "https://wpnews.pro/news/show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos", "markdown": "https://wpnews.pro/news/show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos.md", "text": "https://wpnews.pro/news/show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos.txt", "jsonld": "https://wpnews.pro/news/show-hn-enviouswispr-an-open-source-wispr-flow-alternative-for-macos.jsonld"}}