Challenges / ISMIS 2011 Contest: Music Information Retrieval

Status Closed
Type Scientific
Start 2011-01-10 10:00:00 CET
End 2011-03-21 23:59:59 CET
Prize 2,000$

Registration is required.

Overview

 
 
The challenge is over now. Click here to view the Summary.
 
 
Image

Data mining contest associated with the 19th International Symposium on Methodologies for Intelligent Systems (ISMIS 2011). The task is to recognize different properties of provided music samples, based on extracted sound features. Results of the competition and best solutions will be presented and discussed during ISMIS conference. The contest is held under honorary patronage of the Mazowieckie Province Governor, Jacek Kozłowski.

Image

Internet services expose nowadays vast amounts of multimedia data for exchange and browsing, the most notable example being YouTube. These digital databases cannot be easily searched through, because automatic understanding and indexing of multimedia content is still too difficult for computers - take, for example, the diversity of musical trends and genres, uncommon instruments or the variety of performers and their compositions present in typical multimedia databases. In ISMIS 2011 Contest, we invite all researchers and students interested in sound recognition and related areas (signal processing, data mining, machine learning) to design algorithms for two important and challenging problems of Music Information Retrieval: recognition of music genre (jazz, rock, pop, ...) and recognition of instruments playing together in a given music sample.

The contest comprises 2 tracks:
  • Music Genres - Recognition of genre of music (jazz, rock, pop, ...) from a short sample
  • Music Instruments- Recognition of instruments playing together in a given sample
Tasks are independent. You may participate in both of them or in a selected one.

You may submit solutions many times, for the whole duration of the challenge. They are evaluated instantly after submission and the result is published on Leaderboard, to allow comparison with other participants. However, to avoid overfitting to test data, there are two distinct evaluations: preliminary and final, that employ two separate subsets of test samples. During the contest, Leaderboard presents only preliminary results. Final results are published when the contest is closed and they may differ from preliminary ones. For each task, preliminary samples comprise 35% of test data and final samples - the remaining 65%.

The winning team of the contest track Music Genres as well as the winning team of the contest track Music Instruments will be awarded a prize of 1000 USD each (net, after tax deduction). They will be also invited to prepare papers describing their solutions for inclusion in the ISMIS 2011 proceedings and presentation at the conference. Moreover, top 3 teams on each track will be invited to prepare concise, less technical overviews of their approach, to be published on TunedIT Blog (see this post as an example).

If you have questions please ask them on the Forum. News will be published in the News section, the most important ones will be posted also on Facebook and Twitter.

Schedule

  • January 10, 2011: start of the challenge
  • March 21, 2011: end of the challenge
  • March 28, 2011: deadline for submission of winners' reports for ISMIS
  • April 04, 2011: deadline for submission of descriptions for the Blog
  • June 28-30, 2011: ISMIS conference in Warsaw

Organization

Task 1, Music Genres, was prepared by prof. Bozena Kostek, dr Pawel Zwan, Andrzej Sitek and prof. Andrzej Czyzewski from the Multimedia Systems Department, Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology.

Task 2, Music Instruments, was prepared by prof. Zbigniew Raś and dr Wenxin (Mike) Jiang from University of North Carolina, Charlotte, USA and Warsaw University of Technology, Poland.

The challenge configured and coordinated by Marcin Wojnarski and Joanna Świetlicka, TunedIT. Contact: marcin.wojnarski(AT)tunedit.org.

Media partners

Linux Magazine, Heise-online, Gazeta Studencka, Wiadomości24.pl, 21. Wiek, BI.PL Business Intelligence Portal, Rozpoznawanie Mowy, Laboratorium.

Image

Copyright © 2008-2013 by TunedIT
Design by luksite