Recursive least squares-based online nonnegative matrix factorization (RLS-ONMF), an NMF algorithm based on the RLS method, was developed to solve the NMF problem online. However, this method suffers from a partial-data problem. In this study, the partial-data problem is resolved by developing an improved online NMF algorithm using RLS and a sparsity constraint. The proposed method, RLS-based online sparse NMF (RLS-OSNMF), consists of two steps; an estimation step that optimizes the Euclidean NMF cost function, and a shaping step that satisfies the sparsity constraint. The proposed algorithm was evaluated with recorded speech and music data and with the RWC music database. The results show that the proposed algorithm performs better than conventional RLS-ONMF, especially during the adaptation process.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Seokjin LEE, "RLS-Based On-Line Sparse Nonnegative Matrix Factorization Method for Acoustic Signal Processing Systems" in IEICE TRANSACTIONS on Fundamentals,
vol. E96-A, no. 5, pp. 980-985, May 2013, doi: 10.1587/transfun.E96.A.980.
Abstract: Recursive least squares-based online nonnegative matrix factorization (RLS-ONMF), an NMF algorithm based on the RLS method, was developed to solve the NMF problem online. However, this method suffers from a partial-data problem. In this study, the partial-data problem is resolved by developing an improved online NMF algorithm using RLS and a sparsity constraint. The proposed method, RLS-based online sparse NMF (RLS-OSNMF), consists of two steps; an estimation step that optimizes the Euclidean NMF cost function, and a shaping step that satisfies the sparsity constraint. The proposed algorithm was evaluated with recorded speech and music data and with the RWC music database. The results show that the proposed algorithm performs better than conventional RLS-ONMF, especially during the adaptation process.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.E96.A.980/_p
Copy
@ARTICLE{e96-a_5_980,
author={Seokjin LEE, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={RLS-Based On-Line Sparse Nonnegative Matrix Factorization Method for Acoustic Signal Processing Systems},
year={2013},
volume={E96-A},
number={5},
pages={980-985},
abstract={Recursive least squares-based online nonnegative matrix factorization (RLS-ONMF), an NMF algorithm based on the RLS method, was developed to solve the NMF problem online. However, this method suffers from a partial-data problem. In this study, the partial-data problem is resolved by developing an improved online NMF algorithm using RLS and a sparsity constraint. The proposed method, RLS-based online sparse NMF (RLS-OSNMF), consists of two steps; an estimation step that optimizes the Euclidean NMF cost function, and a shaping step that satisfies the sparsity constraint. The proposed algorithm was evaluated with recorded speech and music data and with the RWC music database. The results show that the proposed algorithm performs better than conventional RLS-ONMF, especially during the adaptation process.},
keywords={},
doi={10.1587/transfun.E96.A.980},
ISSN={1745-1337},
month={May},}
Copy
TY - JOUR
TI - RLS-Based On-Line Sparse Nonnegative Matrix Factorization Method for Acoustic Signal Processing Systems
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 980
EP - 985
AU - Seokjin LEE
PY - 2013
DO - 10.1587/transfun.E96.A.980
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E96-A
IS - 5
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - May 2013
AB - Recursive least squares-based online nonnegative matrix factorization (RLS-ONMF), an NMF algorithm based on the RLS method, was developed to solve the NMF problem online. However, this method suffers from a partial-data problem. In this study, the partial-data problem is resolved by developing an improved online NMF algorithm using RLS and a sparsity constraint. The proposed method, RLS-based online sparse NMF (RLS-OSNMF), consists of two steps; an estimation step that optimizes the Euclidean NMF cost function, and a shaping step that satisfies the sparsity constraint. The proposed algorithm was evaluated with recorded speech and music data and with the RWC music database. The results show that the proposed algorithm performs better than conventional RLS-ONMF, especially during the adaptation process.
ER -