Skip to main content

Installation Guide

OpenWhispr supports macOS, Windows, and Linux. Choose your platform below for detailed installation instructions.
System Requirements:
  • macOS: 10.15 (Catalina) or later, Apple Silicon or Intel
  • Windows: Windows 10 or later, 64-bit
  • Linux: Modern distro with glibc 2.31+ (Ubuntu 20.04+, Fedora 32+, etc.)
  • RAM: 4GB minimum, 8GB recommended for larger models
  • Disk Space: 500MB for app + 75MB to 3GB for local models

Download OpenWhispr

Get the latest release from GitHub:

Download Latest Release

Visit the GitHub Releases page to download installers for all platforms
Current Version: 1.5.5 (actively maintained)

Platform Installation

macOS Installation

1

Choose your architecture

Download the appropriate DMG for your Mac:
  • Apple Silicon (M1, M2, M3, M4): OpenWhispr-{version}-mac-arm64.dmg
  • Intel: OpenWhispr-{version}-mac-x64.dmg
Not sure? Click Apple menu → About This Mac. If you see “Apple M1” or similar, you have Apple Silicon.
2

Install the application

  1. Open the downloaded .dmg file
  2. Drag OpenWhispr.app to your Applications folder
  3. Eject the DMG from Finder
3

First launch (unsigned app)

Since OpenWhispr is not notarized by Apple, you’ll need to bypass Gatekeeper:
  1. Right-click (or Control-click) on OpenWhispr.app in Applications
  2. Select Open from the context menu
  3. Click Open in the security dialog
Do NOT double-click to open the first time. This will show an error. You must right-click → Open to bypass the security warning.
4

Grant microphone permission

When you first start dictation, macOS will prompt for microphone access:
  1. Click OK to allow microphone access
  2. If you accidentally denied it, go to System Settings → Privacy & Security → Microphone
  3. Enable the checkbox for OpenWhispr
5

Grant accessibility permission (optional but recommended)

For automatic text pasting, OpenWhispr needs accessibility permission:
  1. Go to System Settings → Privacy & Security → Accessibility
  2. Click the + button (you may need to unlock with your password)
  3. Navigate to Applications and select OpenWhispr
  4. Enable the checkbox for OpenWhispr
Without accessibility permission, transcribed text copies to clipboard. You’ll need to paste manually with Cmd+V.

macOS-Specific Features

  • Globe Key Support: Optional Fn/Globe key listener for hardware-level dictation trigger
  • Native AppleScript: Reliable text pasting via AppleScript keystroke
  • Menu Bar Integration: Clean menu bar icon with quick access to Control Panel
  • System Settings Links: One-click access to microphone and accessibility settings

Troubleshooting macOS Installation

This happens when trying to double-click on first launch. Solution:
  1. Move OpenWhispr.app to Trash
  2. Re-download and reinstall from DMG
  3. Right-click → Open (not double-click) on first launch
If text still doesn’t paste after granting accessibility:
  1. Remove OpenWhispr from System Settings → Privacy → Accessibility
  2. Quit OpenWhispr completely (Cmd+Q)
  3. Re-add OpenWhispr to accessibility list
  4. Restart the app
Globe key requires Xcode Command Line Tools:
# Install Xcode Command Line Tools
xcode-select --install
After installation, restart OpenWhispr.

Post-Installation Setup

After installing, complete the setup process:
1

Launch OpenWhispr

Open the application from:
  • macOS: Applications folder or Spotlight
  • Windows: Start Menu or Desktop shortcut
  • Linux: Applications menu or terminal (openwhispr)
2

Complete onboarding

Follow the interactive setup wizard:
  1. Choose authentication (cloud sign-in or skip)
  2. Select processing method (cloud, BYOK, or local)
  3. Configure hotkey
  4. Grant system permissions
  5. Test first transcription
See Quick Start Guide for detailed walkthrough.
3

Verify installation

Test that everything works:
  1. Press your hotkey (default: backtick `)
  2. Speak: “This is a test”
  3. Press hotkey again
  4. Verify text appears or copies to clipboard
If the dictation panel appears and records audio, your installation is successful!

Building from Source

For developers or users who want to build OpenWhispr from source:
1

Prerequisites

Install required dependencies:
# Node.js 18+ and npm
# Download from https://nodejs.org

# Verify installation
node --version  # Should be 18.x or higher
npm --version
Platform-specific requirements:
# Xcode Command Line Tools (for Globe key support)
xcode-select --install
2

Clone repository

git clone https://github.com/OpenWhispr/openwhispr.git
cd openwhispr
3

Install dependencies

npm install
4

Download native binaries

# Download whisper.cpp, llama-server, and sherpa-onnx
npm run download:whisper-cpp
npm run download:llama-server
npm run download:sherpa-onnx
5

Build and run

# Run in development mode with hot reload
npm run dev

Updating OpenWhispr

  1. Download latest DMG from GitHub releases
  2. Drag new version to Applications (replace existing)
  3. Launch updated app
Settings and transcription history are preserved during updates.

Uninstalling OpenWhispr

  1. Quit OpenWhispr (Cmd+Q)
  2. Move OpenWhispr.app to Trash from Applications folder
  3. Remove settings (optional):
    rm -rf ~/Library/Application\ Support/OpenWhispr
    rm -rf ~/.cache/openwhispr
    

Next Steps

Quick Start

Complete setup and create your first transcription

Configuration

Customize settings, hotkeys, and processing options

Troubleshooting

Platform-specific issues and solutions

Features

Explore AI processing, custom dictionary, and more

Build docs developers (and LLMs) love