New Algorithms for Non-Negative Matrix Factorization in Applications to Blind Source Separation

Cichocki, Andrzej; Zdunek, Rafał; Амари, Шун-ичи

doi:10.1109/icassp.2006.1661352

Cited by 152 publications

(121 citation statements)

References 3 publications

Supporting

Mentioning

120

Contrasting

Order By: Relevance

“…Kullback-Leibler (KL) divergence between V and W H, denoted D(V ||W H), was found to work well for audio source separation in [2], so we will restrict ourselves to KL divergence in this paper. Generalization to other objective functions using the techniques described in [4] is straightforward.…”

Section: Algorithmmentioning

confidence: 99%

“…During training, we assume availability of a clean speech spectrogram, V speech , of size n f × nst, and a clean (speech-free) noise spectrogram, Vnoise, of size n f × nnt, where n f is the number of frequency bins, nst is the number of speech frames, and nnt is the number of noise frames. Different objective functions lead to different variants of NMF, a number of which are described in [4]. Kullback-Leibler (KL) divergence between V and W H, denoted D(V ||W H), was found to work well for audio source separation in [2], so we will restrict ourselves to KL divergence in this paper.…”

Section: Algorithmmentioning

confidence: 99%

See 1 more Smart Citation

Speech denoising using nonnegative matrix factorization with priors

Wilson¹,

Raj²,

Smaragdis

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

230

137

View full text Add to dashboard Cite

We present a technique for denoising speech using nonnegative matrix factorization (NMF) in combination with statistical speech and noise models. We compare our new technique to standard NMF and to a state-of-the-art Wiener filter implementation and show improvements in speech quality across a range of interfering noise types.This work may not be copied or reproduced in whole or in part for any commercial purpose. Permission to copy in whole or in part without payment of fee is granted for nonprofit educational and research purposes provided that all such whole or partial copies include the following: a notice that such copying is by permission of Mitsubishi Electric Research Laboratories, Inc.; an acknowledgment of the authors and individual contributions to the work; and all applicable portions of the copyright notice. Copying, reproduction, or republishing for any other purpose shall require a license with payment of fee to Mitsubishi Electric Research Laboratories, Inc. All rights reserved. ABSTRACTWe present a technique for denoising speech using nonnegative matrix factorization (NMF) in combination with statistical speech and noise models. We compare our new technique to standard NMF and to a state-of-the-art Wiener filter implementation and show improvements in speech quality across a range of interfering noise types.

show abstract

Section: Algorithmmentioning

confidence: 99%

Section: Algorithmmentioning

confidence: 99%

Speech denoising using nonnegative matrix factorization with priors

Wilson¹,

Raj²,

Smaragdis

et al. 2008

2008 IEEE International Conference on Acoustics, Speech and Signal Processing

230

137

View full text Add to dashboard Cite

show abstract

“…There are several ways to achieve some control of the sparsity. In this paper, we follow the approach proposed in [7] and [9] for KL cost functions, in which the NMF is regularized using non-linear projections based on (3). Applying this procedure, the regularized learning rules are the following,…”

Section: Non-negative Matrix Factorization (Nmf)mentioning

confidence: 99%

“…However, recent NMF-based techniques in speech processing report better results by using NMF with KL divergence [6], [4]. For this reason, in this paper, we propose a NMF-based method for speech denoising which combines the use of the KL divergence with sparseness constraints following the procedure described in [7]. This paper is organized as follows: Section 2 introduces the mathematical background of NMF; in Section 3 we present the speech denoising process using NMF.…”

Section: Introductionmentioning

confidence: 99%

Speech Denoising Using Non-negative Matrix Factorization with Kullback-Leibler Divergence and Sparseness Constraints

Ludeña-Choez

Gallardo-Antolín

2012

Advances in Speech and Language Technologies for Iberian Languages

View full text Add to dashboard Cite

Abstract.A speech denoising method based on Non-Negative Matrix Factorization (NMF) is presented in this paper. With respect to previous related works, this paper makes two contributions. First, our method does not assume a priori knowledge about the nature of the noise. Second, it combines the use of the Kullback-Leibler divergence with sparseness constraints on the activation matrix, improving the performance of similar techniques that minimize the Euclidean distance and/or do not consider any sparsification. We evaluate the proposed method for both, speech enhancement and automatic speech recognitions tasks, and compare it to conventional spectral subtraction, showing improvements in speech quality and recognition accuracy, respectively, for different noisy conditions.

show abstract

“…Atoms within the same group must co-occur by definition. Along with the co-occurrence constraints, to improve the likelihood of co-occurrence within groups, we also impose a smoothness constraint on S using established NMF modifications [7,8]. Fig.…”

Section: That Ssmentioning

confidence: 99%

Multiplicative update rules for nonnegative matrix factorization with co-occurrence constraints

Tjoa

Liu

2010

2010 IEEE International Conference on Acoustics, Speech and Signal Processing

View full text Add to dashboard Cite

Nonnegative matrix factorization (NMF) is a widely-used tool for obtaining low-rank approximations of nonnegative data such as digital images, audio signals, textual data, financial data, and more. One disadvantage of the basic NMF formulation is its inability to control the amount of dependence among the learned dictionary atoms. Enforcing dependence within predetermined groups of atoms allows objects to be represented using multiple atoms instead of only one atom. In this paper, we introduce three simple and convenient multiplicative update rules for NMF that enforce dependence among atoms. Using examples in music transcription, we demonstrate the ability of these updates to represent each musical note with multiple atoms and cluster the atoms for source separation purposes.

show abstract

New Algorithms for Non-Negative Matrix Factorization in Applications to Blind Source Separation

Cited by 152 publications

References 3 publications

Speech denoising using nonnegative matrix factorization with priors

Speech denoising using nonnegative matrix factorization with priors

Speech Denoising Using Non-negative Matrix Factorization with Kullback-Leibler Divergence and Sparseness Constraints

Multiplicative update rules for nonnegative matrix factorization with co-occurrence constraints

Contact Info

Product

Resources

About