VoiceInk is a native macOS application designed for instant, privacy-focused voice-to-text transcription. It targets macOS users seeking an efficient and secure dictation solution, offering offline processing and context-aware AI features.

How It Works

VoiceInk leverages local AI models, specifically mentioning whisper.cpp for high-performance inference of OpenAI's Whisper model. This approach ensures 100% offline processing, prioritizing user privacy. The application aims for near-instantaneous transcription with high accuracy and incorporates "Power Mode" for intelligent, context-aware settings based on the active application or URL.

Quick Start & Requirements

• Install: Download from tryvoiceink.com or install via Homebrew: brew install --cask voiceink.
• Requirements: macOS 14.0 or later.
• Build from Source: Instructions are available in BUILDING.md.
• Documentation: tryvoiceink.com

Highlighted Details

• Local AI models for 99% accurate, instant transcription.
• 100% offline processing for enhanced privacy.
• "Power Mode" for context-aware settings based on active app/URL.
• Personal dictionary for custom terminology and smart text replacements.
• AI Assistant for conversational interactions.

Maintenance & Community

The project is maintained by Pax. Support is available via GitHub issues and Discord (link not provided in README).

Licensing & Compatibility

Licensed under GNU General Public License v3.0 (GPL v3). This is a copyleft license, meaning derivative works must also be licensed under GPL v3. Commercial use or linking with closed-source applications may require careful consideration of GPL v3's terms.

Limitations & Caveats

The project is primarily developed for macOS 14.0+. While open-sourced, the compiled version offers benefits like automatic updates and priority support, suggesting a dual model where core functionality is free but enhanced features are tied to paid licenses.