video2mp3 🎵

A blazingly fast, professional media suite for high-performance transcoding and YouTube downloading.

video2mp3 is an industry-grade media processing suite designed for effortless downloading and high-speed re-encoding. By bridging the raw efficiency of FFmpeg and yt-dlp with a sleek, modern Svelte 5 and Tauri 2.0 interface, it provides a powerful yet intuitive workspace for both single-file tasks and massive batch conversions—all boosted by full hardware acceleration and the bleeding-edge performance of Vite 8 (Rolldown).

🏛️ Project Philosophy

The project is built on the pillars of native performance, professional architecture, and beautiful UI. Recently migrated to the Vite 8 + Svelte 5 stack, video2mp3 leverages the latest in web-native bundling and fine-grained reactivity to deliver an instantaneous user experience.

✨ Key Features

🌍 YouTube & Playlists (Advanced)

• Codec Selection: Force H.264 (compatibility), H.265 (efficiency), or best-quality Remux directly from the UI.
• Smart Staged Workflow: Analyze YouTube URLs in the background while managing your queue.
• Full Playlist Support: Automatically detect and expand entire playlists for batch processing.
• Progress Tracking: Real-time, smooth progress reporting for both download and post-processing phases.

🚀 Hardware Acceleration (Pro Grade)

• Real-time Probing: Dynamically detects available GPU encoders (NVENC, QSV, AMF, VAAPI, VideoToolbox).
• Dynamic Optimization: Automatically configures encoder flags for the best balance between speed and quality.
• Visual Status: Integrated UI tags show exactly which hardware features are currently usable on your system.

🎥 Professional Media Workspace

• Safety First: Integrated Overwrite Protection with confirmation dialogs for existing files.
• Smooth Batching: Overall progress bar uses weighted averages for a fluid, jump-free visual feedback.
• Deep Media Probing: Detailed inspection of containers and codecs (MKV, MP4, etc.) using ffprobe.
• Intelligent Track Selection: Scans all audio streams; automatically pre-selects primary language tracks.

🏗️ Architectural Overview (v2.0.0+)

The project has been refactored into a modern, decoupled web/native structure:

graph TD
    A["Svelte 5 UI (Runes)"] -->|Tauri IPC Commands| B[Tauri Backend - Rust]
    A -->|State Management| C[Reactive Stores]
    B -->|Async Events| A
    B --> D[Core Engine]
    D --> E[FFmpeg/yt-dlp Wrappers]
    D --> F[Hardware Probing]
    D --> G[YAML Configs]

• src/: The entire frontend built with Svelte, Vanilla CSS, and Vite. Handles UI rendering and reactive state.
• src-tauri/src/commands/: Rust endpoints exposed to the frontend via the Tauri IPC bridge.
• src-tauri/src/core/: Pure business logic, hardware detection, and media wrappers.
• src-tauri/src/config/: External YAML profiles defining FFmpeg, yt-dlp, and FFprobe command architectures.

⚙️ Advanced Configuration (YAML)

video2mp3 externalizes its conversion and download logic into YAML configuration files. This allows advanced modifications to FFmpeg parameters without touching the Rust core.

📄 Configuration Files (src-tauri/src/config/)

• ffmpeg.yaml: Defines profiles for audio extraction, remuxing, and hardware-accelerated transcoding.
• ytdlp.yaml: Manages yt-dlp arguments for metadata extraction and various download modes.
• ffprobe.yaml: Configuration for media inspection, duration probing, and stream analysis.

🛠️ Tech Stack

🚀 Getting Started

Prerequisites

• FFmpeg (v5.0+) available in your system's $PATH.
• yt-dlp for YouTube integration features.
• Node.js (v20+) & pnpm.
• Rust Toolchain.

Build from source

1. Clone the repository:

   git clone https://github.com/danloi2/video2mp3.git
   cd video2mp3
   

2. Install frontend dependencies:

   pnpm install
   

3. Run in Development Mode (Hot-Reload):

   pnpm tauri dev
   

4. Build Production Installers:

   pnpm tauri build
   

🤝 Contributing

This project is maintained as a high-quality open-source media suite. Contributions regarding new hardware acceleration profiles or UI refinements are welcome.

Author: Daniel Losada -

⚖️ License

Licensed under the MIT License. See LICENSE for details.