Interactive notebook for parsing and visualising the dataset:
About the dataset: Name identification:
-Chopin Mazurkas: M < opus > - < number >
-PianistID: example "pid1263-01" ->
(from mazurka-discography.txt)
opus |key |performer |year |time |seconds |label |pid |status 6.1 |F# min |Csalog |1996 |3:16 |196 |Hungaroton HCD 31755/6 |1263-01 |-
Folders:
"beat_dyn": One file per Mazurka. Columns:
1. Number of score bar | 2. Number of score beat in bar | 3-x. Performer dynamic values (normalised sones)
each column for a recording identified by PianistID
Rows: 1. PianistID | 2-y. Score beats
"beat_time": One file per Mazurka. Columns:
1. Number of score bar | 2. Number of score beat in bar | 3-x. Performer time values (seconds)
each column for a recording identified by PianistID
Rows: 1. PianistID | 2-y. Score beats
"markings": One file per Mazurka. Rows:
1. Expressive markings found in score | 2. Score beat location of marking
"sones": One file per Mazurka recording
Columns:
1. Time (sec.) | 2. sone value (computed using ma_sone algorithm [1])
[1] www.pampalk.at/ma/documentation.html, accessed 3 January 2018. mazurka-discography.txt from https://charm.kcl.ac.uk/index.html
If using the dataset, please refer to as:
[Katerina Kosta, Oscar F. Bandtlow, and Elaine Chew. "MazurkaBL: Score-aligned loudness, beat, expressive markings data for 2000 Chopin Mazurka recordings”. In Proceedings of the forth International Conference on Technologies for Music Notation and Representation (TENOR). 2018]
Paper: https://drive.google.com/file/d/1h6Ekso91S1U_Ayhx_Y9-NoqV7rxE36ma/view?usp=sharing
Presentation: https://drive.google.com/file/d/1OfrY2jzq_LYt2NnVewM_DVX5ib0-XXBo/view?usp=sharing