Abstract
In this paper, we will test the importance of music complexity
as a factor for melody recognition in multi voiced popular music.
The assumption is that the melody (or lead instrument) will contain the
largest amount of information will be less redundant than the rest.
Measures of melodic complexity of pitch and timing are proposed. We
test the different complexity measures and different prediction strategies,
and evaluate them on the task of predicting which track of a MIDI file
that contains the main melody. Filtering out melody tracks can be useful
when searching large databases for similar songs. 108 melody track
annotated pop songs were included in the experiment.
| Original language | English |
|---|---|
| Title of host publication | Proceedings of the International Workshop on Artificial Intelligenc, 20th International Joint Conference on Artificial Intelligence (IJCAI-07), |
| Number of pages | 9 |
| Publication status | Published - 2007 |
Fields of science
- 102 Computer Sciences
- 102001 Artificial intelligence
- 102003 Image processing
- 102015 Information systems
- 202002 Audiovisual media