Bach in 2014: Music Composition with Recurrent Neural Network

TL;DR

This paper presents a neural network-based framework for music composition, demonstrating that LSTM networks effectively learn musical structure and RProp improves prediction accuracy over traditional methods.

Contribution

Introduces a novel framework combining RProp and LSTM for computer music composition, showing improved learning and prediction of musical structures.

Findings

LSTM networks learn musical structure effectively.

RProp outperforms BPTT in music prediction.

Framework successfully recreates music pieces.

Abstract

We propose a framework for computer music composition that uses resilient propagation (RProp) and long short term memory (LSTM) recurrent neural network. In this paper, we show that LSTM network learns the structure and characteristics of music pieces properly by demonstrating its ability to recreate music. We also show that predicting existing music using RProp outperforms Back propagation through time (BPTT).