Handy

Project Overview

Handy is a free, open-source, and extensible speech-to-text application that works completely offline. Press a shortcut, speak, and have your words appear in any text field. All processing happens on your own computer without sending any information to the cloud.

Key Features

• Configurable keyboard shortcut to start/stop recording (or push-to-talk mode)
• Silence filtering using VAD (Voice Activity Detection) with Silero
• Support for multiple Whisper models (Small/Medium/Turbo/Large) with GPU acceleration
• Support for Parakeet V3 CPU-optimized model with automatic language detection
• Cross-platform: Windows, macOS, and Linux
• Raycast extension integration

Target Audience

• Users needing voice input assistance
• Privacy-conscious users who don't want audio uploaded to the cloud
• Content creators and writers

Quick Start

1. Download from Releases or handy.computer
2. macOS: brew install --cask handy
3. Windows: winget install cjpais.Handy
4. Launch Handy and grant microphone and accessibility permissions
5. Configure shortcuts in Settings and start transcribing

Core Advantages

• Completely free and open-source
• 100% local processing for privacy
• Cross-platform support
• Built with Tauri (Rust + React)

System Requirements

• OS: Windows, macOS, Linux
• Architectures: x86_64, arm64
• Additional Dependencies: None

Notes

• Microphone and accessibility permissions are required
• Homebrew cask and winget packages are not maintained by Handy developers