Onset Detection in Polyphonic Signals by means of Transient Peak Classification - Version 3
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 last years mirex contribution the algorithm has not been modified, but bug fixes and a better parameter optimization procedure should lead to improved performance.
Origine : Fichiers éditeurs autorisés sur une archive ouverte
Loading...