CITATION — REFERENCE ENTRY
Harmonising chorales by probabilistic inference — Advances in Neural Information Processing Systems 17 (NIPS 2004)
- Key
- allan2005-chorale-hmm
- Authors
- Allan, Moray; Williams, Christopher K. I.
- Issued
- 2005
- Type
- paper-conference
- Container
- Advances in Neural Information Processing Systems 17 (NIPS 2004)
- Publisher
- MIT Press
- Pages
- 25-32
Raw CSL JSON
{
"URL": "https://papers.nips.cc/paper/2714-harmonising-chorales-by-probabilistic-inference",
"page": "25-32",
"type": "paper-conference",
"title": "Harmonising chorales by probabilistic inference",
"author": [
{
"given": "Moray",
"family": "Allan"
},
{
"given": "Christopher K. I.",
"family": "Williams"
}
],
"issued": {
"date-parts": [
[
2005
]
]
},
"publisher": "MIT Press",
"container-title": "Advances in Neural Information Processing Systems 17 (NIPS 2004)"
}
Claims
-
Chorale harmonisation, a traditional exercise in Western classical training, asks for three further lines to accompany a given melody. Allan and Williams treated it as a machine learning problem, training hidden Markov models on Bach's chorale harmonisations so that the system learns from examples and can compose new harmonisations.
"We describe how we used a data set of chorale harmonisations composed by Johann Sebastian Bach to train Hidden Markov Models. Using a probabilistic framework allows us to create a harmonisation system which learns from examples, and which can compose new harmonisations. [...] Chorale harmonisation is a traditional part of the theoretical education of Western classical musicians. Given a melody, the task is to create three further lines of music which will sound pleasant when played simultaneously with the original melody."
Available in