Drum extraction in single channel audio signals using multi-layer non negative matrix factor deconvolution

Clément Laroche (1, 2) , Hélène Papadopoulos (1) , Matthieu Kowalski (1, 3) , Gael Richard (2)

1 L2S - Laboratoire des signaux et systèmes
2 Télécom ParisTech
3 PARIETAL - Modelling brain structure, function and variability based on high-field MRI data

In this paper, we propose a supervised multilayer factorization method designed for harmonic/percussive source separation and drum extraction. Our method decomposes the audio signals in sparse orthogonal components which capture the harmonic content, while the drum is represented by an extension of non negative matrix factorization which is able to exploit time-frequency dictionaries to take into account non stationary drum sounds. The drum dictionaries represent various real drum hits and the decomposition has more physical sense and allows for a better interpretation of the results. Experiments on real music data for a harmonic/percussive source separation task show that our method outperforms other state of the art algorithms. Finally, our method is very robust to non stationary harmonic sources that are usually poorly decomposed by existing methods.

Mots clés: Index Terms— Drum extraction; Source separation; Non- negative matrix factorization

Domaines: Traitement du signal et de l'image [eess.SP]

Clément Laroche, Hélène Papadopoulos, Matthieu Kowalski, Gael Richard. Drum extraction in single channel audio signals using multi-layer non negative matrix factor deconvolution. The 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017), Mar 2017, Nouvelle Orleans, United States. ⟨10.1109/icassp.2017.7952115⟩. ⟨hal-01438851⟩