Skip to Main content Skip to Navigation

Modèles d'instruments pour la séparation de sources et la transcription d'enregistrements musicaux

Abstract : For about fifteen years the study of chamber music recordings has focused on two distinct viewpoints : source separation and polyphonic transcription. Source separation tries to extract from a recording the signals corresponding to each musical instrument playing. Polyphonic transcription aims at describing a recording by a set of parameters : names of the instruments, pitch and loudness of the notes, etc. Existing methods, based on spatial and spectro-temporal analysis of the recordings, provide satisfying results in simple cases. But their performance generally degrades fast with more instruments than a fixed limit, under reverberant conditions, with instruments of similar playing ranges or with notes at harmonic intervals. Our hypothesis is that these methods often suffer from too generic models of instrumental sources. We propose to address this by creating specific instrument models based on learning. In this dissertation, we justify this hypothesis by studying the relevant information present in musical recordings and its use by existing methods. Then we describe new probabilistic instrument models inspired from Independent Subspace Analysis (ISA) and we give a few examples of learnt instruments. Finally we exploit these models to separate and transcribe realistic recordings, among which CD tracks and synthetic convolutive or underdetermined mixtures of these tracks.
Complete list of metadatas

Cited literature [127 references]  Display  Hide  Download
Contributor : Emmanuel Vincent <>
Submitted on : Wednesday, December 8, 2010 - 5:02:18 PM
Last modification on : Friday, March 22, 2019 - 1:31:46 AM
Document(s) archivé(s) le : Thursday, March 10, 2011 - 12:37:51 PM


  • HAL Id : tel-00544710, version 1


Emmanuel Vincent. Modèles d'instruments pour la séparation de sources et la transcription d'enregistrements musicaux. Informatique [cs]. Université Pierre et Marie Curie - Paris VI, 2004. Français. ⟨tel-00544710⟩



Record views


Files downloads