Student's T nonnegative matrix factorization and positive semidefinite tensor factorization for single-channel audio source separation

Yoshii, Kazuyoshi; Itoyama, Katsutoshi; Goto, Masataka

doi:10.1109/icassp.2016.7471635

Cited by 44 publications

(46 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is equivalent to the model of t-PSDTF [18] with the observation yjn. By applying the t-PSDTF algorithm to (23), we derive the update rules as follows: To avoid the ambiguity of the scales of v kjn and U kn , we adjust the scales at each iteration so that tr (U kn ) = 1.…”

Section: Update Of Source Model V Kjn and U Knmentioning

confidence: 99%

Convergence-Guaranteed Independent Positive Semidefinite Tensor Analysis Based on Student’s T Distribution

Kondo

Fukushige

Takamune

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

In this paper, we address a blind source separation (BSS) problem and propose a new extended framework of independent positive semidefinite tensor analysis (IPSDTA). IPSDTA is a state-of-the-art BSS method that enables us to take interfrequency correlations into account, but the generative model is limited within the multivariate Gaussian distribution and its parameter optimization algorithm does not guarantee stable convergence. To resolve these problems, first, we propose to extend the generative model to a parametric multivariate Student's t distribution that can deal with various types of signal. Secondly, we derive a new parameter optimization algorithm that guarantees the monotonic nonincrease in the cost function, providing stable convergence. Experimental results reveal that the cost function in the conventional IPSDTA does not display monotonically nonincreasing properties. On the other hand, the proposed method guarantees the monotonic nonincrease in the cost function and outperforms the conventional ILRMA and IPSDTA in the source-separation performance.Index Terms-blind source separation, independent positive semidefinite tensor analysis, Student's t distribution

show abstract

Section: Update Of Source Model V Kjn and U Knmentioning

confidence: 99%

Convergence-Guaranteed Independent Positive Semidefinite Tensor Analysis Based on Student’s T Distribution

Kondo

Fukushige

Takamune

et al. 2020

ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

View full text Add to dashboard Cite

show abstract

“…Machine‐learning techniques have attracted considerable interest recently, with efforts to improve BSS performance and address the underdetermined source separation problem in practical situations where the number of sound sources is greater than the number of channels for the observed signals (i.e., the number of microphones). Underdetermined BSS techniques have been widely studied too, not only for speech separation but also for music separation, as music signals are usually provided as a stereo channel signal consisting of multiple instrument and vocal sounds .…”

Section: Introductionmentioning

confidence: 99%

Environmental sound processing and its applications

Miyazaki

Toda

Hayashi

et al. 2019

IEEJ Transactions Elec Engng

View full text Add to dashboard Cite

As part of the effort to develop techniques for understanding environments using sound, many studies in the field of computational auditory scene analysis have focused on using computers to perform functions carried out naturally by the human auditory system. Thanks to recent progress in machine‐learning techniques, these environmental sound‐processing techniques have significantly improved and a widening variety of applications has resulted in considerable interest in this field. In this review, we introduce the fundamental techniques of environmental sound processing, as well as recent advances in front‐end and back‐end processing and potential applications for these techniques. Prospects for further progress in the field of environmental sound processing and the challenges still to be overcome are also discussed. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

show abstract

“…Moreover, a good initialization in a Gaussian NMF model is crucial to avoid staying stuck in a local minimum [3]. Many studies in the single-channel case have shown a better robustness to initialization when the signal is modeled in the TF domain with as heavy tail distribution [22,19].…”

Section: Introductionmentioning

confidence: 99%