Stochastic approximation and a nonlocally weighted soft-constrained recursive algorithm for blind separation of reverberant speech mixtures

Previous Article
Analytical proof of space-time chaos in Ginzburg-Landau equations
DCDS Home
This Issue
Next Article

December 2010, 28(4): 1753-1767. doi: 10.3934/dcds.2010.28.1753

Stochastic approximation and a nonlocally weighted soft-constrained recursive algorithm for blind separation of reverberant speech mixtures

1.	Department of Mathematics, University of California, Irvine, CA 92697, United States
2.	Department of Mathematics/Mathematics, University of California at Irvine, Irvine, CA 92697

Received October 2009 Revised February 2010 Published June 2010

Abstract
Related Papers

We review statistical equations for blind source separation problems, then introduce their stochastic approximation and recursive algorithms. The recurrence resembles discretization of nonlinear systems of ordinary differential equations which may not have global solutions in general. Though scaling variables were used before to control finite time blowup, instabilities may arise from small divisor problem during silent periods of speech signals, and asymptotic balance as a necessary condition for convergence was ignored. To resolve these deficiencies, we propose a nonlocally weighted soft-constrained recursive algorithm. The nonlocal weighting of the iterations promotes stability and convergence of the algorithm. The scaling variables evolve by soft-constrained difference equations. Computations on synthetic speech mixtures based on measured binaural room impulse responses in enclosed rooms with reverberation time up to 1 second show that the new algorithm achieves consistently higher signal-to-interference ratio improvement than existing methods. The algorithm is observed to be stable and convergent, and is applied to separation of room recorded mixtures of song and music as well.

Keywords: reverberant speech mixtures, weighted constrained recurrence, Stochastic approximation, blind separation..

Mathematics Subject Classification: Primary: 94A12, 65H10, 65C60; Secondary: 60G3.

Citation: Meng Yu, Jack Xin. Stochastic approximation and a nonlocally weighted soft-constrained recursive algorithm for blind separation of reverberant speech mixtures. Discrete & Continuous Dynamical Systems - A, 2010, 28 (4) : 1753-1767. doi: 10.3934/dcds.2010.28.1753

[1]	Yuanchang Sun, Lisa M. Wingen, Barbara J. Finlayson-Pitts, Jack Xin. A semi-blind source separation method for differential optical absorption spectroscopy of atmospheric gas mixtures. Inverse Problems & Imaging, 2014, 8 (2) : 587-610. doi: 10.3934/ipi.2014.8.587
[2]	Igor E. Pritsker and Richard S. Varga. Weighted polynomial approximation in the complex plane. Electronic Research Announcements, 1997, 3: 38-44.
[3]	Jiawei Chen, Shengjie Li, Jen-Chih Yao. Vector-valued separation functions and constrained vector optimization problems: optimality and saddle points. Journal of Industrial & Management Optimization, 2017, 13 (5) : 1-18. doi: 10.3934/jimo.2018174
[4]	Nicolas Lecoq, Helena Zapolsky, P.K. Galenko. Numerical approximation of the Chan-Hillard equation with memory effects in the dynamics of phase separation. Conference Publications, 2011, 2011 (Special) : 953-962. doi: 10.3934/proc.2011.2011.953
[5]	Lili Chang, Wei Gong, Guiquan Sun, Ningning Yan. PDE-constrained optimal control approach for the approximation of an inverse Cauchy problem. Inverse Problems & Imaging, 2015, 9 (3) : 791-814. doi: 10.3934/ipi.2015.9.791
[6]	Nikolai Dokuchaev. On strong causal binomial approximation for stochastic processes. Discrete & Continuous Dynamical Systems - B, 2014, 19 (6) : 1549-1562. doi: 10.3934/dcdsb.2014.19.1549
[7]	Xiaojun Li, Xiliang Li, Kening Lu. Random attractors for stochastic parabolic equations with additive noise in weighted spaces. Communications on Pure & Applied Analysis, 2018, 17 (3) : 729-749. doi: 10.3934/cpaa.2018038
[8]	Lee DeVille, Nicole Riemer, Matthew West. Convergence of a generalized Weighted Flow Algorithm for stochastic particle coagulation. Journal of Computational Dynamics, 2018, 0 (0) : 1-26. doi: 10.3934/jcd.2019003
[9]	Ardeshir Ahmadi, Hamed Davari-Ardakani. A multistage stochastic programming framework for cardinality constrained portfolio optimization. Numerical Algebra, Control & Optimization, 2017, 7 (3) : 359-377. doi: 10.3934/naco.2017023
[10]	Beatris Adriana Escobedo-Trujillo, Alejandro Alaffita-Hernández, Raquiel López-Martínez. Constrained stochastic differential games with additive structure: Average and discount payoffs. Journal of Dynamics & Games, 2018, 5 (2) : 109-141. doi: 10.3934/jdg.2018008
[11]	Mou-Hsiung Chang, Tao Pang, Moustapha Pemy. Finite difference approximation for stochastic optimal stopping problems with delays. Journal of Industrial & Management Optimization, 2008, 4 (2) : 227-246. doi: 10.3934/jimo.2008.4.227
[12]	Xiaojun Chen, Guihua Lin. CVaR-based formulation and approximation method for stochastic variational inequalities. Numerical Algebra, Control & Optimization, 2011, 1 (1) : 35-48. doi: 10.3934/naco.2011.1.35
[13]	Yanfeng Guo, Jinqiao Duan, Donglong Li. Approximation of random invariant manifolds for a stochastic Swift-Hohenberg equation. Discrete & Continuous Dynamical Systems - S, 2016, 9 (6) : 1701-1715. doi: 10.3934/dcdss.2016071
[14]	Ralf Banisch, Carsten Hartmann. A sparse Markov chain approximation of LQ-type stochastic control problems. Mathematical Control & Related Fields, 2016, 6 (3) : 363-389. doi: 10.3934/mcrf.2016007
[15]	Hui-Qiang Ma, Nan-Jing Huang. Neural network smoothing approximation method for stochastic variational inequality problems. Journal of Industrial & Management Optimization, 2015, 11 (2) : 645-660. doi: 10.3934/jimo.2015.11.645
[16]	Yongchao Liu, Hailin Sun, Huifu Xu. An approximation scheme for stochastic programs with second order dominance constraints. Numerical Algebra, Control & Optimization, 2016, 6 (4) : 473-490. doi: 10.3934/naco.2016021
[17]	Marvin S. Müller. Approximation of the interface condition for stochastic Stefan-type problems. Discrete & Continuous Dynamical Systems - B, 2017, 22 (11) : 1-23. doi: 10.3934/dcdsb.2019121
[18]	Angelo Morro. Nonlinear diffusion equations in fluid mixtures. Evolution Equations & Control Theory, 2016, 5 (3) : 431-448. doi: 10.3934/eect.2016012
[19]	Dingshi Li, Lin Shi, Xiaohu Wang. Long term behavior of stochastic discrete complex Ginzburg-Landau equations with time delays in weighted spaces. Discrete & Continuous Dynamical Systems - B, 2017, 22 (11) : 1-28. doi: 10.3934/dcdsb.2019046
[20]	Yong Ren, Huijin Yang, Wensheng Yin. Weighted exponential stability of stochastic coupled systems on networks with delay driven by $ G $-Brownian motion. Discrete & Continuous Dynamical Systems - B, 2019, 24 (7) : 3379-3393. doi: 10.3934/dcdsb.2018325

2018 Impact Factor: 1.143

American Institute of Mathematical Sciences