Adaptation of MVSep-MDX23 algorithm for Colab, with few tweaks:

https://colab.research.google.com/github/jarredou/MVSEP-MDX23-Colab_v2/blob/v2.3/MVSep-MDX23-Colab.ipynb

Recent changes:

v2.3

• HQ3-Instr model replaced by VitLarge23 (thanks to MVSep)
• Improved MDXv2 processing (thanks to Anjok)
• Improved BigShifts algo (v2)
• BigShifts processing added to MDXv3 & VitLarge
• Faster folder batch processing

Original work:

Model based on Demucs4, MDX neural net architectures and some MDX weights from Ultimate Vocal Remover project (thanks Kimberley Jensen for great high quality vocal models). Brought to you by MVSep.com.

    python inference.py --input_audio mixture1.wav mixture2.wav --output_folder ./results/

With this command audios with names "mixture1.wav" and "mixture2.wav" will be processed and results will be stored in ./results/ folder in WAV format.

• Note 1: If you have not enough GPU memory you can use CPU (--cpu), but it will be slow. Additionally you can use single ONNX (--single_onnx), but it will decrease quality a little bit. Also reduce of chunk size can help (--chunk_size 200000).
• Note 2: In current revision code requires less GPU memory, but it process multiple files slower. If you want old fast method use argument --large_gpu. It will require > 11 GB of GPU memory, but will work faster.

Quality comparison with best separation models performed on MultiSong Dataset.

• Note: SDR - signal to distortion ratio. Larger is better.

• Script for GUI (based on PyQt5): gui.py.
• You can download standalone program for Windows here: zip1, zip2. Unzip archives and to start program double click run.bat.
• Program will download all needed neural net models from internet at the first run.
• GUI supports Drag & Drop of multiple files.
• Progress bar available.

• arxiv paper

@misc{solovyev2023benchmarks,
      title={Benchmarks and leaderboards for sound demixing tasks}, 
      author={Roman Solovyev and Alexander Stempkovskiy and Tatiana Habruseva},
      year={2023},
      eprint={2305.07489},
      archivePrefix={arXiv},
      primaryClass={cs.SD}
}