CITATION — REFERENCE ENTRY
Automatic melody harmonization with triad chords: A comparative study — arXiv
- Key
- yeh2021-melody-harmonization
- Authors
- Yeh, Yin-Cheng; Hsiao, Wen-Yi; Fukayama, Satoru; Kitahara, Tetsuro; Genchel, Benjamin; Liu, Hao-Min; Dong, Hao-Wen; Chen, Yian; Leong, Terence; Yang, Yi-Hsuan
- Issued
- 2021-4-27
- Type
- article
- Container
- arXiv
Raw CSL JSON
{
"URL": "https://arxiv.org/abs/2001.02360",
"type": "article",
"title": "Automatic melody harmonization with triad chords: A comparative study",
"author": [
{
"given": "Yin-Cheng",
"family": "Yeh"
},
{
"given": "Wen-Yi",
"family": "Hsiao"
},
{
"given": "Satoru",
"family": "Fukayama"
},
{
"given": "Tetsuro",
"family": "Kitahara"
},
{
"given": "Benjamin",
"family": "Genchel"
},
{
"given": "Hao-Min",
"family": "Liu"
},
{
"given": "Hao-Wen",
"family": "Dong"
},
{
"given": "Yian",
"family": "Chen"
},
{
"given": "Terence",
"family": "Leong"
},
{
"given": "Yi-Hsuan",
"family": "Yang"
}
],
"issued": {
"date-parts": [
[
2021,
4,
27
]
]
},
"number": "2001.02360v3",
"container-title": "arXiv"
}
Claims
-
Automatic melody harmonization generates a sequence of chords to accompany a given melody; it is challenging because a melody can be harmonized in many ways and what is pleasant is subjective and genre-dependent. Yeh and colleagues compared template-matching, hidden Markov model, genetic algorithm, and deep learning approaches on 9,226 melody/chord pairs with up to 48 triad chords, with a listening study of 202 participants. Deep learning models generally performed better; a model accounting for chord function performed best and did not differ significantly from human-composed progressions in interestingness or overall rating, while the genetic algorithm was the most interesting of the other automatic methods.
"the task of automatic melody harmonization, in which a model aims to generate a sequence of chords to serve as the harmonic accompaniment of a given multiple-bar melody sequence. [...] including a template matching based model, a hidden Markov based model, a genetic algorithm based model, and two deep learning based models. The evaluation is conducted on a dataset of 9,226 melody/chord pairs we newly collect for this study, considering up to 48 triad chords [...] and a subjective study with 202 participants. [...] Melody harmonization is a challenging task as there are multiple ways to harmonize the same melody; what makes a particular harmonization pleasant is subjective, and often dependent on musical genre and other contextual factors. [...] For Interestingness, GA performs the best among the five automatic methods [...] there is significant performance difference between the MTHarmonzer progressions and the original human-composed progressions in terms of Coherence and Chord Progression [...], but no significant difference in terms of Interestingness and Overall. [...] Our evaluation shows that deep learning models indeed perform better than non-deep learning ones in a variety of aspects, including harmonicity and interestingness. Moreover, a deep learning model that takes the function of chords into account reaches the best result among the evaluated models."
Available in