Center for Machine Learning and Intelligent Systems
About  Citation Policy  Donate a Data Set  Contact


Repository Web            Google
View ALL Data Sets

× Check out the beta version of the new UCI Machine Learning Repository we are currently testing! Contact us if you have any issues, questions, or concerns. Click here to try out the new site.

Bach Choral Harmony Data Set
Download: Data Folder, Data Set Description

Abstract: The data set is composed of 60 chorales (5665 events) by J.S. Bach (1675-1750). Each event of each chorale is labelled using 1 among 101 chord labels and described through 14 features.

Data Set Characteristics:  

Sequential

Number of Instances:

5665

Area:

N/A

Attribute Characteristics:

N/A

Number of Attributes:

17

Date Donated

2014-05-20

Associated Tasks:

Classification

Missing Values?

N/A

Number of Web Hits:

54605


Source:

-- Creators: Daniele P. Radicioni and Roberto Esposito
-- Donor: Daniele P. Radicioni (radicion '@' di.unito.it) and Roberto Esposito (esposito '@' di.unito.it)
-- Date: May, 2014


Data Set Information:

Pitch classes information has been extracted from MIDI sources downloaded
from (JSB Chorales)[[Web Link]]. Meter information has
been computed through the Meter program which is part of the Melisma
music analyser (Melisma)[[Web Link]].
Chord labels have been manually annotated by a human expert.


Attribute Information:

1. Choral ID: corresponding to the file names from (Bach Central)[[Web Link]].
2. Event number: index (starting from 1) of the event inside the chorale.
3-14. Pitch classes: YES/NO depending on whether a given pitch is present.
Pitch classes/attribute correspondence is as follows:
C -> 3
C#/Db -> 4
D -> 5
...
B -> 14
15. Bass: Pitch class of the bass note
16. Meter: integers from 1 to 5. Lower numbers denote less accented events,
higher numbers denote more accented events.
17. Chord label: Chord resonating during the given event.


Relevant Papers:

1. D. P. Radicioni and R. Esposito. Advances in Music Information Retrieval,
chapter BREVE: an HMPerceptron-Based Chord Recognition System. Studies
in Computational Intelligence, Zbigniew W. Ras and Alicja Wieczorkowska
(Editors), Springer, 2010.
2. Esposito, R. and Radicioni, D. P., CarpeDiem: Optimizing the Viterbi
Algorithm and Applications to Supervised Sequential Learning, Journal
of Machine Learning Research, 10(Aug):1851-1880, 2009.



Citation Request:

D. P. Radicioni and R. Esposito. Advances in Music Information Retrieval, chapter BREVE: an HMPerceptron-Based Chord Recognition System. Studies in Computational Intelligence, Zbigniew W. Ras and Alicja Wieczorkowska (Editors), Springer, 2010.


Supported By:

 In Collaboration With:

About  ||  Citation Policy  ||  Donation Policy  ||  Contact  ||  CML