M. Lee and K.-M. Sung, “A Mixing Matrix Estimation Method for Blind Source Separation of Underdetermined Audio Mixture,” in Proc. AES Convention 133, Oct. 2012, pp. 8798:1-10. [Online]. Available: https://aes.org/publications/elibrary-page/?id=16540
Lee M, Sung KM. A Mixing Matrix Estimation Method for Blind Source Separation of Underdetermined Audio Mixture. In: AES Convention 133. Audio Engineering Society; 2012.pp. 8798:1-10. Available from: https://aes.org/publications/elibrary-page/?id=16540
@inproceedings{Lee2012_16540,
author = {Lee, Mingu and Sung, Keong-Mo},
title = {{A Mixing Matrix Estimation Method for Blind Source Separation of Underdetermined Audio Mixture}},
booktitle = {AES Convention 133},
pages = {8798:1-10},
url = {https://aes.org/publications/elibrary-page/?id=16540},
year = {2012},
month = oct,
publisher = {Audio Engineering Society},
address = {San Francisco}
}
TY - CPAPER
TI - A Mixing Matrix Estimation Method for Blind Source Separation of Underdetermined Audio Mixture
AU - Lee, Mingu
AU - Sung, Keong-Mo
T2 - AES Convention 133
SP - 1
EP - 10
M1 - Paper 8798
PY - 2012
DA - 2012/10/06
UR - https://aes.org/publications/elibrary-page/?id=16540
PB - Audio Engineering Society
CY - San Francisco
LA - en
AB - A new mixing matrix estimation method for under-determined blind source separation of audio signals is proposed. By statistically modeling the local features, i.e., the magnitude ratio and phase difference of the mixtures, in a time-frequency region, a region can have information of the mixing angle of a source with reliability amounted to its likelihood. Regional data are then clustered with statistical tests based on their likelihood to produce estimates for the mixing angle of the sources as well as the number of them. Experimental results show that the proposed mixing matrix estimation algorithm outperform the existing methods.
ER -