Searching For Music Mixing Graphs: A Pruning Approach

Sungho Lee1*, Marco A. Martínez-Ramírez2, Wei-Hsiang Liao2, Stefan Uhlich3, Giorgio Fabbro3, Kyogu Lee1, and Yuki Mitsufuji2,4
1Seoul National University, 2Sony AI, 3Sony Europe B.V, 4Sony Group Corporation
*Work done during an internship at Sony AI

Submitted to DAFx 2024
Preprint

Supplementary documents
Appendix and combined (main + appendix)

Code (to be released)


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.

Framework-train

Music mixing graph search via iterative pruning. We start from a mixing console $G_\mathrm{c}$ and optimize its parameters $\mathbf{P}_\mathrm{c}$. Then, we alternate between pruning and fine-tuning stages, obtaining a sparse graph $G_\mathrm{p}$ and its parameters $\mathbf{P}_\mathrm{p}$ that does not degrade the match qualiy of the mixing console up to a tolerance threshold $\tau$.

Framework-train

Finding a sparse graph $G_\mathrm{p}$ from a differentiable mixing console $G_\mathrm{c}$. i: input, o: output, m: mix, e: equalizer, c: compressor, n: noisegate, s: stereo imager, g: gain/panning, d: multitap delay, and r: reverb.


Audio Samples

Torres - NewSkin
Mid Error Side Error
Target mix
Base graph
+ Gain/panning
+ Stereo imager
Equalizer
+ Reverb
+ Compressor
+ Noisegate
+ Multitap delay
Pruned, 0.1
Pruned, 0.01
Pruned, 0.001


AlexanderRoss - VelvetCurtain
Mid Error Side Error
Target mix
Base graph
+ Gain/panning
+ Stereo imager
Equalizer
+ Reverb
+ Compressor
+ Noisegate
+ Multitap delay
Pruned, 0.1
Pruned, 0.01
Pruned, 0.001


AClassicEducation - NightOwl
Mid Error Side Error
Target mix
Base graph
+ Gain/panning
+ Stereo imager
Equalizer
+ Reverb
+ Compressor
+ Noisegate
+ Multitap delay
Pruned, 0.1
Pruned, 0.01
Pruned, 0.001


Cayetana - MissThing
Mid Error Side Error
Target mix
Base graph
+ Gain/panning
+ Stereo imager
Equalizer
+ Reverb
+ Compressor
+ Noisegate
+ Multitap delay
Pruned, 0.1
Pruned, 0.01
Pruned, 0.001


MusicDelta - Britpop
Mid Error Side Error
Target mix
Base graph
+ Gain/panning
+ Stereo imager
Equalizer
+ Reverb
+ Compressor
+ Noisegate
+ Multitap delay
Pruned, 0.1
Pruned, 0.01
Pruned, 0.001