Reading the repo

Opening the box…

How to set up openwhispr (OpenWhispr/openwhispr) · gitmanual

Assembly instructions

OpenWhispr/openwhispr

Voice-to-text dictation app with local (Nvidia Parakeet/Whisper) and cloud models (BYOK). Privacy-first and available cross-platform.

Node.jsVite#ai#anthropic#cross-platform#gemini

Difficulty: Moderate
Time: about 15 min
Stars: ★ 10,882
Activity: Updated this month
License: MIT
Open issues: 469
Got it running: Be the first
Looked up: 44 min ago

Saved setup from 44 min ago

Start here

Website ↗openwhispr.comThe project's own site.Docs ↗docs.openwhispr.comLinked from the README.

Will it work on my device?

Yes, it installs on your macOS

A ready-made macOS download is in the latest release.

Checking for

macOS Windows Linux

Download for macOS

1.11.0 · 10/29/2026 ↗

OpenWhispr-1.11.0-arm64.dmg

dmg · Apple Silicon · 343.3 MB · 5,021 downloads · best match

Download

Drag the app into Applications.

curl -LO "https://github.com/OpenWhispr/openwhispr/releases/download/v1.11.0/OpenWhispr-1.11.0-arm64.dmg"
open "OpenWhispr-1.11.0-arm64.dmg"

Other macOS downloads (3)

OpenWhispr-1.11.0.dmg

dmg · 341.7 MB · 779 downloads

Download

Drag the app into Applications.

curl -LO "https://github.com/OpenWhispr/openwhispr/releases/download/v1.11.0/OpenWhispr-1.11.0.dmg"
open "OpenWhispr-1.11.0.dmg"

OpenWhispr-1.11.0-arm64-mac.zip

zip · Apple Silicon · 341.9 MB · 6,941 downloads

Download

If it contains an .app, move it to Applications.

curl -LO "https://github.com/OpenWhispr/openwhispr/releases/download/v1.11.0/OpenWhispr-1.11.0-arm64-mac.zip"
unzip "OpenWhispr-1.11.0-arm64-mac.zip"

OpenWhispr-1.11.0-mac.zip

zip · 340.5 MB · 331 downloads

Download

If it contains an .app, move it to Applications.

curl -LO "https://github.com/OpenWhispr/openwhispr/releases/download/v1.11.0/OpenWhispr-1.11.0-mac.zip"
unzip "OpenWhispr-1.11.0-mac.zip"

Install with a package manager

Homebrew

brew install --cask openwhispr

From the README, exactly as the maintainers wrote it.

Did you get it running?

No reports for macOS yet.

Still stuck? Ask an AI

Opens a chat with this manual already filled in, for macOS. It uses your own account; gitmanual sends nothing.

Ask Claude Ask ChatGPT

Parts list

1×Node.js 24runtime
1×npm package manager

24× settings in .env

VITE_AUTH_URLoptional
VITE_OPENWHISPR_API_URLoptional
OPENAI_API_KEYsecret
WHISPER_MODELoptional
LANGUAGEoptional
UI_LANGUAGEoptional
ANTHROPIC_API_KEYsecretoptional
GEMINI_API_KEYsecretoptional
GROQ_API_KEYsecretoptional
TINFOIL_API_KEYsecretoptional
MISTRAL_API_KEYsecretoptional
OPENWHISPR_LOG_LEVELoptional
EXPO_PUBLIC_API_URL
EXPO_PUBLIC_SENTRY_DSNoptional
EXPO_PUBLIC_APPSFLYER_DEV_KEYsecretoptional
EXPO_PUBLIC_SUPERWALL_IOS_API_KEYsecret
EXPO_PUBLIC_SUPERWALL_ANDROID_API_KEYsecret
EXPO_PUBLIC_REVENUECAT_IOS_API_KEYsecret
EXPO_PUBLIC_REVENUECAT_ANDROID_API_KEYsecret
EXPO_PUBLIC_AFFILIATE_DOMAINoptional
EXPO_PUBLIC_AFFILIATE_PUBLISHABLE_KEYsecretoptional
EXPO_PUBLIC_GOOGLE_CALENDAR_IOS_CLIENT_IDoptional
EXPO_PUBLIC_GOOGLE_CALENDAR_IOS_CLIENT_ID_DEVoptional
EXPO_PUBLIC_KEYBOARD_AUDIO_FORMATsecret

Share this manual

Is this your repo?

Add a badge to your README so newcomers find these steps.

[![Setup manual](https://gitmanual.com/OpenWhispr/openwhispr/badge.svg)](https://gitmanual.com/OpenWhispr/openwhispr)

Good to know

The README mentions NVIDIA CUDA. Check its hardware section: some features (or all of them) may need an NVIDIA GPU.
Has CLAUDE.md — instructions for coding agents. You can open Claude Code (or another agent) in the cloned folder and ask it to do the setup for you.
Nested projects: native/meeting-aec-helper/CMakeLists.txt, openwhispr-mobile/Gemfile, openwhispr-mobile/modules/app-group-storage/package.json, openwhispr-mobile/modules/apple-llm/package.json and 9 more

package.json scripts (76)

compile:globenode scripts/build-globe-listener.js
compile:fast-pastenode scripts/build-macos-fast-paste.js
compile:winkeysnode scripts/build-windows-key-listener.js
compile:linuxkeysnode scripts/build-linux-key-listener.js
compile:text-monitornode scripts/build-text-monitor.js
compile:winpastenode scripts/build-windows-fast-paste.js
compile:linux-pastenode scripts/build-linux-fast-paste.js
compile:linux-system-audionode scripts/build-linux-system-audio.js
compile:media-remotenode scripts/build-media-remote.js
compile:mediaremote-adapternode scripts/build-mediaremote-adapter.js
compile:mic-listenernode scripts/build-macos-mic-listener.js
compile:calendar-listenernode scripts/build-macos-calendar-listener.js
compile:audio-tapnode scripts/build-macos-audio-tap.js
compile:window-boundsnode scripts/build-macos-window-bounds.js
compile:nativenpm run compile:globe && npm run compile:fast-paste && npm run compile:winkeys && npm run compile:linuxkeys && npm run compile:winpaste && npm run compile:linux-paste && npm run compile:linux-system-audio && npm run compile:text-monitor && npm run compile:media-remote && npm run compile:mediaremote-adapter && npm run compile:mic-listener && npm run compile:calendar-listener && npm run compile:audio-tap && npm run compile:window-bounds
prestartnpm run compile:native && npm run download:meeting-aec-helper && npm run download:sherpa-onnx && npm run download:yt-dlp && npm run download:qdrant && npm run download:embedding-model && npm run download:whisper-vad-model
startelectron .
devconcurrently -k -r "npm:dev:renderer" "npm:dev:main"
predev:mainnpm run compile:native && npm run download:meeting-aec-helper && npm run download:sherpa-onnx && npm run download:yt-dlp && npm run download:qdrant && npm run download:embedding-model && npm run download:whisper-vad-model && npm run download:brand-fonts
dev:maincross-env NODE_ENV=development node scripts/run-electron.js --dev
dev:renderercd src && vite
prebuildnpm run compile:native && npm run download:whisper-cpp && npm run download:llama-server && npm run download:sherpa-onnx && npm run download:yt-dlp && npm run download:qdrant && npm run download:meeting-aec-helper && npm run download:whisper-vad-model && npm run download:embedding-model -- --for-build && npm run download:diarization-models -- --output-dir resources/bin/diarization-models && npm run download:brand-fonts
buildcd src && vite build && cd .. && electron-builder
postinstallelectron-builder install-app-deps
build:renderercd src && vite build

Files we read (12)

.env.example

.github/CONTRIBUTING.md

.github/workflows/build-and-notarize.yml

.github/workflows/build-linux-text-monitor.yml

.github/workflows/build-meeting-aec-helper.yml

.github/workflows/build-windows-fast-paste.yml

.github/workflows/build-windows-key-listener.yml

.github/workflows/build-windows-mic-listener.yml

.nvmrc

README.md

openwhispr-mobile/.env.example

package.json

Branch main · 3,705 files in the repo