Music databases are useful tools for the basic research on the musical structure, classification and other such large-scale analytic and empirical techniques. Databases also facilitate the development of technical applications, such as various musical search and query methods (e.g., query-by-humming or finding melodic using symbolic information) that all fall under the general term of Music Information Retrieval (MIR, see ISMIR, The International Conferences on Music Information Retrieval and Related Activities).
There are not many large scale folk song collections available for research purposes. The Essen Folk song collection (Schaffrath, 1995) is one good example and various other, less well annotated databases exist as well, e.g., Themefinder and Meldex.
The Finnish folk song database is available for research purposes (that is, noncommercial) provided that the attribution is maintained, i.e., you must give the original authors credit. To cite the Finnish Folk Music Database, use: Eerola, T., & Toiviainen, P. (2004). Suomen Kansan eSävelmät. Finnish Folk Song Database. [11.3.2004]. Available: http://www.jyu.fi/musica/sks/.
The format of the Tune Index is shown below. Most of the columns are self-explanatory. Collection refers to abbreviated names of the subcollections, i.e., LS1= Laulusävelmät 1, RS1= Runosävelmät 1 and so on, see info page. The cadence columns refers to the classification system proposed by Ilmari Krohn (1902-1903).
Update 6.2.2006: Note velocities set to uniform value (100) and file version updated for Matlab 7.
Update 17.4.2006: Tune index updated (53 tunes had wrong time-signatures and 166 songs were labeled by a single time-signature although they contained multiple time-signatures. These both types of mistakes have now been corrected by TE).
Copyright of the original collection by © Finnish Literary Society, based on the publication Suomen Kansan Sävelmiä (Edited by I. Krohn, 1898-1930).
Copyright of the electronic collection: © Tuomas Eerola & Petri Toiviainen.
Visitors since 12/2004: