hhSCNet

SCNet in an easier-to-install package and with some small changes. The result should be easier to integrate into your Python package.

See SCNet: Sparse Compression Network for Music Source Separation

Usage

Command Line Interface

Separate audio

hhSCNet inference --pathInput "./input/" --pathOutput "./output/" --modelConfiguration "./conf/config.yaml" --checkpoint "./result/checkpoint.th"

Train a model

hhSCNet train --modelConfiguration "./conf/config.yaml" --pathSave "./result/"

Python API

Inference

from hhSCNet import runInference

runInference(
    pathInput="./input/",
    pathOutput="./output/",
    modelConfiguration="./conf/config.yaml",
    checkpoint="./result/checkpoint.th"
)

Training

from hhSCNet import trainModel

trainModel(
    modelConfiguration="./conf/config.yaml",
    pathSave="./result/"
)

Install this package

From Github

pip install hhSCNet@git+https://github.com/hunterhogan/hhSCNet.git

From a local directory

Windows

git clone https://github.com/hunterhogan/hhSCNet.git \path\to\hhSCNet
pip install hhSCNet@file:\path\to\hhSCNet

POSIX

git clone https://github.com/hunterhogan/hhSCNet.git /path/to/hhSCNet
pip install hhSCNet@file:/path/to/hhSCNet

Install updates

pip install --upgrade hhSCNet@git+https://github.com/hunterhogan/hhSCNet.git

Cite the original paper

@misc{tong2024scnet,
      title={SCNet: Sparse Compression Network for Music Source Separation}, 
      author={Weinan Tong and Jiaxu Zhu and Jun Chen and Shiyin Kang and Tao Jiang and Yang Li and Zhiyong Wu and Helen Meng},
      year={2024},
      eprint={2401.13276},
      archivePrefix={arXiv},
      primaryClass={eess.AS}
}