VoiceInk

Project Overview

VoiceInk is a native macOS voice-to-text application that transcribes what you say to text almost instantly. After 5 months of development, the author decided to open-source it, aiming to make it the most efficient and privacy-focused voice-to-text solution for macOS.

Key Features

• Accurate transcription: Local AI models with 99% accuracy, almost instant
• Privacy first: 100% offline processing, data never leaves your device
• Smart modes: Auto-applies perfect pre-configured settings based on the app/URL
• Context aware: Smart AI that understands your screen content and adapts
• Global shortcuts: Configurable keyboard shortcuts for recording and push-to-talk
• Personal dictionary: Train the AI to understand your unique terminology
• AI Assistant: Built-in voice assistant mode for conversational interaction

Target Audience

• macOS users who need efficient voice-to-text functionality
• Privacy-conscious users who want offline voice processing
• Users looking for an open-source alternative to Superwhisper and Wispr Flow

Quick Start

1. Download the free trial from tryvoiceink.com
2. Or install via Homebrew: brew install --cask voiceink
3. Build from source: follow BUILDING.md

Core Advantages

• Fully open-source under GPL v3.0
• 100% offline processing for privacy
• Smart context awareness and auto-mode switching

System Requirements

• OS: macOS 14.4 or later