Vox Documentation

Fast, accurate voice transcription with AI enhancement for macOS and Windows.

What is Vox?

Vox is a desktop application for macOS and Windows that transcribes your speech into text. It uses Whisper, an advanced open-source speech recognition technology from OpenAI, running entirely on your device. All processing happens locally, ensuring your privacy while providing fast and accurate transcriptions.

Key Features

• Local Processing - Speech recognition runs entirely on your device
• AI Enhancement - Optional post-processing with AWS Bedrock, DeepSeek, or custom LLM providers
• Always Available - System tray integration with customizable keyboard shortcuts
• Smart Dictionary - Teach Vox custom words and phrases for better accuracy
• Multi-Language - Support for English, Portuguese, Spanish, French, German, and more
• Privacy First - Your audio never leaves your device

Quick Start

1. Download Vox from the official website
2. Install and launch the application
3. Grant required permissions (Microphone, Accessibility, Keychain)
4. Download a speech model (Accurate recommended for most users)
5. Start recording with ⌘ + Space (hold) or ⌘ + ⌥ + Space (toggle)

New to Vox?

Follow our Getting Started guide for detailed setup instructions.

Documentation

📦 Setup

• Getting Started
• System Permissions

⚙️ Configuration

• General Settings
• Speech Models
• AI Enhancement
• Keyboard Shortcuts

📝 Usage

• Recording Audio
• Transcriptions
• Custom Dictionary

Need Help?

• Issues & Bugs: GitHub Issues
• Discussions: GitHub Discussions
• Source Code: GitHub Repository