CITATION — REFERENCE ENTRY
HMM-Based Chord Recognition — FMP Notebooks
- Key
- fmp-c5-chord-hmm
- Authors
- Müller, Meinard; Weiß, Christof
- Type
- webpage
- Container
- FMP Notebooks
- Publisher
- International Audio Laboratories Erlangen
Raw CSL JSON
{
"URL": "https://www.audiolabs-erlangen.de/resources/MIR/FMP/C5/C5S3_ChordRec_HMM.html",
"type": "webpage",
"title": "HMM-Based Chord Recognition",
"author": [
{
"given": "Meinard",
"family": "Müller"
},
{
"given": "Christof",
"family": "Weiß"
}
],
"accessed": {
"date-parts": [
[
2026,
11,
10
]
]
},
"publisher": "International Audio Laboratories Erlangen",
"container-title": "FMP Notebooks"
}
Claims
-
A simple template-based chord recognizer produces many random-like chord changes.
"This leads to many random-like chord changes when using a simple template-based chord recognizer."
-
Hidden Markov models improve on frame-by-frame template matching through their transition model: with high self-transition probabilities, the recognizer tends to stay on the current chord, which smooths the output.
"In the case of high self-transition probabilities, a chord recognizer tends to stay in the current chord rather than change to another one, which can be regarded as a kind of smoothing."
Available in