Summary

Grably is a free, open-source, privacy-first desktop application designed for downloading and transcribing web media. It targets users who require a secure, offline solution for content acquisition and analysis, offering a significant benefit of ad-free, tracking-free operation with fast, local processing capabilities. The project aims to provide a reliable tool for media management without compromising user privacy.

How It Works

This native macOS application is built using Rust and the Tauri framework, ensuring high performance, cross-platform potential, and minimal resource usage. Media downloading is handled by the robust yt-dlp tool, which supports over 1000 websites and offers granular control over download quality (up to 4K) and batch processing for playlists. For transcription, Grably integrates Whisper.cpp, enabling powerful, offline AI-powered speech-to-text conversion. Key transcription features include speaker diarization, multi-language support, and export to common formats like SRT and VTT.

Quick Start & Requirements

The most straightforward method to begin using Grably is by downloading the pre-built Universal binary (.dmg) for macOS directly from the project's Releases page on GitHub. For users wishing to build from source, prerequisites include Node.js (version 18 or higher), Rust (latest stable version), and macOS Command Line Tools. The application automatically manages and bundles necessary external binaries like yt-dlp, ffmpeg, and Whisper.cpp via a script.

Highlighted Details

• Universal Media Downloader: Capable of downloading video and audio from an extensive list of over 1000 websites, including popular platforms like YouTube, Instagram, TikTok, and Twitter. It supports selecting desired video/audio quality options and facilitates playlist or batch downloads.
• AI-Powered Transcription: Leverages OpenAI's Whisper AI model via Whisper.cpp for accurate, offline transcription of audio and video content. Features include automatic speaker detection and diarization, support for multiple languages, and flexible export options to TXT, SRT, and VTT formats.
• Privacy-First Design: A core tenet of Grably is its commitment to user privacy, achieved through 100% offline processing where no user data ever leaves the device. The application is fully open-source, allowing for code audits, and is explicitly free from analytics, tracking, or telemetry.
• Performance & Efficiency: Developed with Rust and Tauri for a native macOS experience, Grably boasts lightning-fast download speeds through parallel processing. It is designed for minimal resource consumption, resulting in a small application size (approximately 200MB) and efficient operation on Apple Silicon and Intel Macs.

Maintenance & Community

User support, bug reporting, and feature requests are primarily managed through the project's GitHub repository, specifically via GitHub Issues and Discussions. The project welcomes community contributions. For direct inquiries, an email address (support@grably.space) is provided. Future development plans include expanding platform support to Windows and Linux.

Licensing & Compatibility

Grably is distributed under the permissive MIT License. This license permits broad usage, modification, and distribution, making it highly compatible with commercial applications and closed-source projects without significant restrictions.

Limitations & Caveats

The current stable release of Grably is exclusively available for macOS. While support for Windows and Linux platforms is actively planned and indicated as "coming soon," these operating systems are not yet supported in the current version.