Results 1  10
of
41
C.: Multichannel nonnegative matrix factorization in convolutive mixtures for audio source separation
 IEEE Trans. Audio, Speech, Language Process
, 2010
"... We consider inference in a general datadriven objectbased model of multichannel audio data, assumed generated as a possibly underdetermined convolutive mixture of source signals. Each source is given a model inspired from nonnegative matrix factorization (NMF) with the ItakuraSaito divergence, wh ..."
Abstract

Cited by 78 (17 self)
 Add to MetaCart
(Show Context)
We consider inference in a general datadriven objectbased model of multichannel audio data, assumed generated as a possibly underdetermined convolutive mixture of source signals. Each source is given a model inspired from nonnegative matrix factorization (NMF) with the ItakuraSaito divergence, which underlies a statistical model of superimposed Gaussian components. We address estimation of the mixing and source parameters using two methods. The first one consists of maximizing the exact joint likelihood of the multichannel data using an expectationmaximization algorithm. The second method consists of maximizing the sum of individual likelihoods of all channels using a multiplicative update algorithm inspired from NMF methodology. Our decomposition algorithms were applied to stereo music and assessed in terms of blind source separation performance. Index Terms — Multichannel audio, nonnegative matrix factorization, nonnegative tensor factorization, underdetermined convolutive blind source separation. 1.
A general flexible framework for the handling of prior information in audio source separation
 IEEE Transactions on Audio, Speech and Signal Processing
, 2012
"... Abstract—Most of audio source separation methods are developed for a particular scenario characterized by the number of sources and channels and the characteristics of the sources and the mixing process. In this paper we introduce a general audio source separation framework based on a library of str ..."
Abstract

Cited by 46 (16 self)
 Add to MetaCart
(Show Context)
Abstract—Most of audio source separation methods are developed for a particular scenario characterized by the number of sources and channels and the characteristics of the sources and the mixing process. In this paper we introduce a general audio source separation framework based on a library of structured source models that enable the incorporation of prior knowledge about each source via userspecifiable constraints. While this framework generalizes several existing audio source separation methods, it also allows to imagine and implement new efficient methods that were not yet reported in the literature. We first introduce the framework by describing the model structure and constraints, explaining its generality, and summarizing its algorithmic implementation using a generalized expectationmaximization algorithm. Finally, we illustrate the abovementioned capabilities of the framework by applying it in several new and existing configurations to different source separation problems. We have released a software tool named Flexible Audio Source Separation Toolbox (FASST) implementing a baseline version of the framework in Matlab. Index Terms—Audio source separation, local Gaussian model, nonnegative matrix factorization, expectationmaximization I.
The 2008 signal separation evaluation campaign: A communitybased approach to largescale evaluation
 in ICA, 2009
"... Abstract. This paper introduces the first communitybased Signal Separation Evaluation Campaign (SiSEC 2008), coordinated by the authors. This initiative aims to evaluate source separation systems following specifications agreed between the entrants. Four speech and music datasets were contributed, ..."
Abstract

Cited by 38 (12 self)
 Add to MetaCart
(Show Context)
Abstract. This paper introduces the first communitybased Signal Separation Evaluation Campaign (SiSEC 2008), coordinated by the authors. This initiative aims to evaluate source separation systems following specifications agreed between the entrants. Four speech and music datasets were contributed, including synthetic mixtures as well as microphone recordings and professional mixtures. The source separation problem was split into four tasks, each evaluated via different objective performance criteria. We provide an overview of these datasets, tasks and criteria, summarize the results achieved by the submitted systems and discuss organization strategies for future campaigns. 1
First Stereo Audio Source Separation Evaluation Campaign: Data, algorithms and results
 in: Proc. 7th Int. Conf. on Independent Component Analysis and Signal Separation
, 2007
"... ..."
(Show Context)
R.: Underdetermined instantaneous audio source separation via local Gaussian modeling
 In: Proc. 8th Int. Conf. on Independent Component Analysis and Signal Separation. (2009) 775–782
"... Abstract. Underdetermined source separation is often carried out by modeling timefrequency source coefficients via a fixed sparse prior. This approach fails when the number of active sources in one timefrequency bin is larger than the number of channels or when active sources lie on both sides of ..."
Abstract

Cited by 16 (10 self)
 Add to MetaCart
(Show Context)
Abstract. Underdetermined source separation is often carried out by modeling timefrequency source coefficients via a fixed sparse prior. This approach fails when the number of active sources in one timefrequency bin is larger than the number of channels or when active sources lie on both sides of an inactive source. In this article, we partially address these issues by modeling timefrequency source coefficients via Gaussian priors with free variances. We study the resulting maximum likelihood criterion and derive a fast noniterative optimization algorithm that finds the global minimum. We show that this algorithm outperforms stateoftheart approaches over stereo instantaneous speech mixtures. 1
Twomicrophone separation of speech mixtures
 IEEE Transactions on Neural Networks
, 2008
"... ..."
(Show Context)
NONNEGATIVE MATRIX FACTORIZATION AND SPATIAL COVARIANCE MODEL FOR UNDERDETERMINED REVERBERANT AUDIO SOURCE SEPARATION
"... We address the problem of blind audio source separation in the underdetermined and convolutive case. The contribution of each source to the mixture channels in the timefrequency domain is modeled by a zeromean Gaussian random vector with a full rank covariance matrix composed of two terms: a vari ..."
Abstract

Cited by 15 (6 self)
 Add to MetaCart
(Show Context)
We address the problem of blind audio source separation in the underdetermined and convolutive case. The contribution of each source to the mixture channels in the timefrequency domain is modeled by a zeromean Gaussian random vector with a full rank covariance matrix composed of two terms: a variance which represents the spectral properties of the source and which is modeled by a nonnegative matrix factorization (NMF) model and another full rank covariance matrix which encodes the spatial properties of the source contribution in the mixture. We address the estimation of these parameters by maximizing the likelihood of the mixture using an expectationmaximization (EM) algorithm. Theoretical propositions are corroborated by experimental studies on stereo reverberant music mixtures. 1.
F.: Blind spectralGMM estimation for underdetermined instantaneous audio source separation
 In: Proc. ICA. (2009
"... Abstract. The underdetermined blind audio source separation problem is often addressed in the timefrequency domain by assuming that each timefrequency point is an independently distributed random variable. Other approaches which are not blind assume a more structured model, like the Spectral Gauss ..."
Abstract

Cited by 13 (10 self)
 Add to MetaCart
(Show Context)
Abstract. The underdetermined blind audio source separation problem is often addressed in the timefrequency domain by assuming that each timefrequency point is an independently distributed random variable. Other approaches which are not blind assume a more structured model, like the Spectral Gaussian Mixture Models (SpectralGMMs), thus exploiting statistical diversity of audio sources in the separation process. However, in this last approach, SpectralGMMs are supposed to be learned from some training signals. In this paper, we propose a new approach for learning SpectralGMMs of the sources without the need of using training signals. The proposed blind method significantly outperforms stateoftheart approaches on stereophonic instantaneous music mixtures. 1
Source number estimation and clustering for underdetermined blind source separation
 in International Workshop on Acoustic Echo and Noise Control (IWAENC
, 2008
"... Much research has been undertaken in the field of blind source separation (BSS) and a large number of algorithms have been developed. However, most of them assume that the number of sources is known. In this paper we present an algorithm to estimate the number of sources in the (over)determined a ..."
Abstract

Cited by 7 (0 self)
 Add to MetaCart
(Show Context)
Much research has been undertaken in the field of blind source separation (BSS) and a large number of algorithms have been developed. However, most of them assume that the number of sources is known. In this paper we present an algorithm to estimate the number of sources in the (over)determined and underdetermined case. We call this algorithm NOSET (Number of Sources Estimation Technique). We start from a description of the BSS problem, give a short overview of the socalled observation vector clustering algorithm and then present our approach. It is based on directionofarrival (DOA) estimation from reliable timefrequency points and a clustering of the DOA estimates. The estimated DOAs can be used to recover the source signals by performing a nearestneighbor classification of the observation vectors instead of the conventional kmeans clustering procedure which is sensitive to the choice of initial centroids. 1.
Double Sparsity: Towards Blind Estimation of Multiple Channels
"... Abstract. We propose a framework for blind multiple filter estimation from convolutive mixtures, exploiting the timedomain sparsity of the mixing filters and the disjointness of the sources in the timefrequency domain. The proposed framework includes two steps: (a) a clustering step, to determine ..."
Abstract

Cited by 6 (1 self)
 Add to MetaCart
(Show Context)
Abstract. We propose a framework for blind multiple filter estimation from convolutive mixtures, exploiting the timedomain sparsity of the mixing filters and the disjointness of the sources in the timefrequency domain. The proposed framework includes two steps: (a) a clustering step, to determine the frequencies where each source is active alone; (b) a filter estimation step, to recover the filter associated to each source from the corresponding incomplete frequency information. We show how to solve the filter estimation step (b) using convex programming, and we explore numerically the factors that drive its performance. Step (a) remains challenging, and we discuss possible strategies that will be studied in future work. Key words: blind filter estimation, sparsity, convex optimisation 1