- Recherche et sélection de publications
|
Genre specific dictionaries for harmonic/percussive source separation
- Clément Laroche #1 #2, Hélène Papadopoulos #1, Matthieu Kowalski #1, Gaël Richard #2
-
#1 |
Laboratoire des signaux et systèmes (L2S)
- UMR8506 CNRS
- SUPELEC
- Univ Paris-Sud
|
#2 |
Télécom ParisTech
|
- References
- ISMIR, New York, USA, September 2016,
- Abstract
Blind source separation usually obtains limited performance on real and polyphonic music signals. To overcome these limitations, it is common to rely on prior knowledge under the form of side information as in \emph{Informed Source Separation} or on machine learning paradigms applied on a training database. In the context of source separation based on factorization models such as the \emph{Non-negative Matrix Factorization}, this supervision can be introduced by learning specific dictionaries. However, due to the large diversity of musical signals it is not easy to build sufficiently compact and precise dictionaries that will well characterize the large array of audio sources. In this paper, we argue that it is relevant to construct genre-specific dictionaries. Indeed, we show on a task of harmonic/percussive source separation that the dictionaries built on genre-specific training subsets yield better performances than cross-genre dictionaries.
- Keywords
- Category
- Paper in proceedings
- Research Area(s)
- Computer Science/Signal and Image Processing
- Identifier(s)
-
Bibliographic key laroche2016genre
- File(s)
-
- Export
-
- Last update
- on january 18, 2017 by Clément Laroche
|