Music Composition with Deep Learning: A Review
link.springer.com/chapter/10.1007/978-3-031-18444-4_2
1 Users
0 Comments
9 Highlights
0 Notes
Tags
Top Highlights
1980s
early 1980s such as the Experiments in Musical Intelligence (EMI) [12] by David Cope from 1983 to 1989 or Analogiques A and B by Iannis Xenakis
David Cope proposed the combination of Markov chains with grammars for automatic music composition, and other relevant works such as Project1 (PR1) by Koening [2] were born
techniques can be grouped in the field of algorithmic music composition which is a way of composing by means of formalizable methods
type of composing consists of a controlled procedure which is based on mathematical instructions that must be followed in a fixed order.
There are several methods inside the algorithmic composition such as Markov models, generative grammars, cellular automata, genetic algorithms, transition networks, or chaos theory
these techniques and other probabilistic methods are combined with deep neural networks (NNs) in order to condition them or help them to better model music which is the case of DeepBach
DL models for music generation normally use NN architectures that are proven to perform well in other fields such as computer vision or natural language processing (NLP)
pre-trained models in these fields that can be used for music generation. This is called transfer learning
Ready to highlight and find good content?
Glasp is a social web highlighter that people can highlight and organize quotes and thoughts from the web, and access other like-minded people’s learning.