Wykorzystanie sztucznej inteligencji do generowania tre\'sci muzycznych

TL;DR

This thesis presents a novel method for generating short musical phrases using a DCGAN trained on MIDI data converted into piano roll images, achieving musically interesting results.

Contribution

It introduces a new approach of translating MIDI data into images for GAN training, enhancing music generation quality and diversity.

Findings

Generated musical phrases contain interesting rhythmic and harmonic structures.

The image-based approach improves the quality of generated music.

The method shows promising results compared to existing solutions.

Abstract

This thesis is presenting a method for generating short musical phrases using a deep convolutional generative adversarial network (DCGAN). To train neural network were used datasets of classical and jazz music MIDI recordings. Our approach introduces translating the MIDI data into graphical images in a piano roll format suitable for the network input size, using the RGB channels as additional information carriers for improved performance. The network has learned to generate images that are indistinguishable from the input data and, when translated back to MIDI and played back, include several musically interesting rhythmic and harmonic structures. The results of the conducted experiments are described and discussed, with conclusions for further work and a short comparison with selected existing solutions.