onset detection by means of transient peak classification
Résumé
The extended abstract describes an onset detection algorithm that is based on a classification of spectral peaks into transient and non-transient peaks and a statistical model of the classification results to prevent detection of random transient peaks due to noise. Compared to the version used for MIREX 2010 the algorithm presented here differs with respect to the treatment of soft harmoinc onsets.
Origine | Fichiers produits par l'(les) auteur(s) |
---|
Loading...