Searching For Music Mixing Graphs: A Pruning Approach

TL;DR

This paper introduces a differentiable pruning method to reverse engineer and simplify music mixing graphs, resulting in sparse models that match the quality of full mixing consoles and can aid neural network training.

Contribution

It presents a novel differentiable pruning approach to extract sparse, effective music mixing graphs from complex processing chains.

Findings

Sparse mixing graphs match full console quality

Method generalizes across various datasets

Graphs can be used for neural network training

Abstract

Music mixing is compositional -- experts combine multiple audio processors to achieve a cohesive mix from dry source tracks. We propose a method to reverse engineer this process from the input and output audio. First, we create a mixing console that applies all available processors to every chain. Then, after the initial console parameter optimization, we alternate between removing redundant processors and fine-tuning. We achieve this through differentiable implementation of both processors and pruning. Consequently, we find a sparse mixing graph that achieves nearly identical matching quality of the full mixing console. We apply this procedure to dry-mix pairs from various datasets and collect graphs that also can be used to train neural networks for music mixing applications.