GRAFX: An Open-Source Library for Audio Processing Graphs in PyTorch
Department of Intelligence and Information, Seoul National University, Seoul, South Korea
rXiv:2408.03204 [cs.SD], (6 Aug 2024)
@misc{lee2024grafxopensourcelibraryaudio,
title={GRAFX: An Open-Source Library for Audio Processing Graphs in PyTorch},
author={Sungho Lee and Marco Martínez-Ramírez and Wei-Hsiang Liao and Stefan Uhlich and Giorgio Fabbro and Kyogu Lee and Yuki Mitsufuji},
year={2024},
eprint={2408.03204},
archivePrefix={arXiv},
primaryClass={cs.SD},
url={https://arxiv.org/abs/2408.03204}
}
We present GRAFX, an open-source library designed for handling audio processing graphs in PyTorch. Along with various library functionalities, we describe technical details on the efficient parallel computation of input graphs, signals, and processor parameters in GPU. Then, we show its example use under a music mixing scenario, where parameters of every differentiable processor in a large graph are optimized via gradient descent. The code is available.
August 18, 2024 by hgpu
