I want to share my latest open source project with you. It is an AI powered tool to find out what genre a song is. Its name is Song genre detector. In short SGD. It is Windows only.
SGD Portable is a perfect tool for musicians like me who are unsure how to categorize their tracks. There is a youtube release video available where i lead you through the steps.
Important License Information:The tool itself is Freeware and Open Source under an MIT License. You are free to use, modify, and distribute it for personal projects.
However, the underlying AI model (Maest) operates under a Non-Commercial License. This means you cannot use the genre results of this tool in a commercial environment or for profit-generating work. For hobbyists, students, and private musicians, it is completely free to use.
You can find the tool at Github: https://github.com/ReinerBforartists/sgd-portable
And of course at my page: https://haizytiles.reinerstilesets.de/i ... -portable/
Key Features:
Drag & Drop analysis for single or multiple songs at once.
Automatic genre detection with visual chart summaries.
Tag Cloud generation based on detected genres.
Customizable genre list via Python configuration file.
Modern UI built with Gradio (Python/PyTorch backend).
Free and open source under an MIT license.
Installation Guide:
Download the appropriate ZIP file from the link above.
Unpack the folder on your computer.
Run start.bat.
Wait for the model download to finish (approx. 2 GB for CPU version, approx. 6 GB for CUDA/GPU version). This is a one-time action; future starts will be instant.
The tool opens automatically in your browser. It runs portably inside a virtual environment. To uninstall, simply delete the folder.
Technical Details:
UI Framework: Gradio
Backend: Python, PyTorch
AI Model: Maest (Selected for best performance over self-trained models)
Portable and runs in its own Virtual Environment (Venv)
If you encounter any bugs or have suggestions, please report them via the GitHub Repository.
Have fun with the tool - Tiles