Algorithmic Composition of Melodies with Deep Recurrent Neural Networks
This addresses the problem of automated music composition for creators and researchers, but it is incremental as it builds on existing neural network methods for sequence modeling.
The paper tackled the challenge of algorithmic composition by training gated recurrent unit networks on a large corpus of melodies to generate new melodies or continuations that are coherent with the learned style, achieving results that produce interesting and coherent outputs.
A big challenge in algorithmic composition is to devise a model that is both easily trainable and able to reproduce the long-range temporal dependencies typical of music. Here we investigate how artificial neural networks can be trained on a large corpus of melodies and turned into automated music composers able to generate new melodies coherent with the style they have been trained on. We employ gated recurrent unit networks that have been shown to be particularly efficient in learning complex sequential activations with arbitrary long time lags. Our model processes rhythm and melody in parallel while modeling the relation between these two features. Using such an approach, we were able to generate interesting complete melodies or suggest possible continuations of a melody fragment that is coherent with the characteristics of the fragment itself.