Signal decomposition by means of classification of spectral peaks

Axel Roebel (1) , Miroslav Zivanovic (1) , Xavier Rodet (1)

1 Analyse et synthèse sonores [Paris]

In extending previous work on detecting transient spectral peaks we here investigate into the distinction between sinusoidal and noise components by means of classification of spectral peaks. The classification is based on descriptors derived from properties related to time-frequency distributions. In contrast to existing methods, the descriptors are designed to properly deal with non-stationary sinusoids, which considerably increases the range of applications. The experimental investigation shows superior classification results compared to the standard correlation-based approach.

Mots clés: Signal decomposition; classification of spectral peaks; sinusoids; noise.

Mots clés: Informatique musicale

Domaines: Acoustique [physics.class-ph]; Traitement du signal et de l'image [eess.SP]; Neurosciences

Axel Roebel, Miroslav Zivanovic, Xavier Rodet. Signal decomposition by means of classification of spectral peaks. International Computer Music Conference (ICMC), Nov 2004, Miami, United States. pp.446-449. ⟨hal-01161169⟩