Latent Space Oddity: Exploring Latent Spaces to Design Guitar Timbres

TL;DR

This paper presents a new neural network architecture with an interpretable latent space that enables musicians to design novel guitar timbres by combining features of existing amplifiers.

Contribution

It introduces a convolutional network with a domain-informed latent space for intuitive guitar tone synthesis and design.

Findings

The model can generate diverse guitar timbres by manipulating latent variables.

It allows intuitive combination and subtraction of amplifier characteristics.

The approach demonstrates effective control over guitar tone synthesis.

Abstract

We introduce a novel convolutional network architecture with an interpretable latent space for modeling guitar amplifiers. Leveraging domain knowledge of popular amplifiers spanning a range of styles, the proposed system intuitively combines or subtracts characteristics of different amplifiers, allowing musicians to design entirely new guitar timbres.