Abstract
We describe variable markov models we have used for statistical learning of musical sequences, then we present the factor oracle, a data structure proposed by Crochemore & al for string matching. We show the relation between this structure and the previous models and indicate how it can be adapted for learning musical sequences and generating improvisations in a real-time context.
Original language | English |
---|---|
Pages (from-to) | 604-610 |
Number of pages | 7 |
Journal | Soft Computing |
Volume | 8 |
Issue number | 9 |
DOIs | |
State | Published - 1 Dec 2004 |
Externally published | Yes |
Keywords
- Computer music
- Factor oracle
- Improvisation
- Incremental parsing
- Machine learning
- Prediction suffix trees
- Suffix trees
- Variable markov models
ASJC Scopus subject areas
- Theoretical Computer Science
- Software
- Geometry and Topology