Skip to Main content Skip to Navigation
Conference papers

Genre specific dictionaries for harmonic/percussive source separation

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 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 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.
Document type :
Conference papers
Complete list of metadata

Cited literature [39 references]  Display  Hide  Download
Contributor : Clément Laroche <>
Submitted on : Monday, October 23, 2017 - 9:08:26 AM
Last modification on : Wednesday, October 14, 2020 - 4:20:27 AM
Long-term archiving on: : Wednesday, January 24, 2018 - 1:14:00 PM


Files produced by the author(s)


  • HAL Id : hal-01353252, version 2


Clément Laroche, Hélène Papadopoulos, Matthieu Kowalski, Gael Richard. Genre specific dictionaries for harmonic/percussive source separation. ISMIR 2016 - The 17th International Society for Music Information Retrieval Conference, Aug 2016, New York, United States. ⟨hal-01353252v2⟩



Record views


Files downloads