Progressive Generative Adversarial Binary Networks for Music Generation

TL;DR

This paper introduces a progressive GAN training approach for music generation, expanding model complexity over time and incorporating binary neurons to produce binary-valued musical outputs, resulting in improved stability and output quality.

Contribution

It presents a novel progressive training method for GANs in music generation combined with deterministic binary neurons for binary output, enhancing stability and output clarity.

Findings

Progressive training stabilizes GAN training for music.

Binary neurons produce clear binary-valued musical outputs.

Model achieves larger tensor sizes with improved quality.

Abstract

Recent improvements in generative adversarial network (GAN) training techniques prove that progressively training a GAN drastically stabilizes the training and improves the quality of outputs produced. Adding layers after the previous ones have converged has proven to help in better overall convergence and stability of the model as well as reducing the training time by a sufficient amount. Thus we use this training technique to train the model progressively in the time and pitch domain i.e. starting from a very small time value and pitch range we gradually expand the matrix sizes until the end result is a completely trained model giving outputs having tensor sizes [4 (bar) x 96 (time steps) x 84 (pitch values) x 8 (tracks)]. As proven in previously proposed models deterministic binary neurons also help in improving the results. Thus we make use of a layer of deterministic binary neurons at the end of the generator to get binary valued outputs instead of fractional values existing between 0 and 1.