This paper describes a real-time blind source separation (BSS) method for moving speech signals in a room. Our method employs frequency domain independent component analysis (ICA) using a blockwise batch algorithm in the first stage, and the separated signals are refined by postprocessing using crosstalk component estimation and non-stationary spectral subtraction in the second stage. The blockwise batch algorithm achieves better performance than an online algorithm when sources are fixed, and the postprocessing compensates for performance degradation caused by source movement. Experimental results using speech signals recorded in a real room show that the proposed method realizes robust real-time separation for moving sources. Our method is implemented on a standard PC and works in realtime.
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
Ryo MUKAI, Hiroshi SAWADA, Shoko ARAKI, Shoji MAKINO, "Blind Source Separation for Moving Speech Signals Using Blockwise ICA and Residual Crosstalk Subtraction" in IEICE TRANSACTIONS on Fundamentals,
vol. E87-A, no. 8, pp. 1941-1948, August 2004, doi: .
Abstract: This paper describes a real-time blind source separation (BSS) method for moving speech signals in a room. Our method employs frequency domain independent component analysis (ICA) using a blockwise batch algorithm in the first stage, and the separated signals are refined by postprocessing using crosstalk component estimation and non-stationary spectral subtraction in the second stage. The blockwise batch algorithm achieves better performance than an online algorithm when sources are fixed, and the postprocessing compensates for performance degradation caused by source movement. Experimental results using speech signals recorded in a real room show that the proposed method realizes robust real-time separation for moving sources. Our method is implemented on a standard PC and works in realtime.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/e87-a_8_1941/_p
Copy
@ARTICLE{e87-a_8_1941,
author={Ryo MUKAI, Hiroshi SAWADA, Shoko ARAKI, Shoji MAKINO, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Blind Source Separation for Moving Speech Signals Using Blockwise ICA and Residual Crosstalk Subtraction},
year={2004},
volume={E87-A},
number={8},
pages={1941-1948},
abstract={This paper describes a real-time blind source separation (BSS) method for moving speech signals in a room. Our method employs frequency domain independent component analysis (ICA) using a blockwise batch algorithm in the first stage, and the separated signals are refined by postprocessing using crosstalk component estimation and non-stationary spectral subtraction in the second stage. The blockwise batch algorithm achieves better performance than an online algorithm when sources are fixed, and the postprocessing compensates for performance degradation caused by source movement. Experimental results using speech signals recorded in a real room show that the proposed method realizes robust real-time separation for moving sources. Our method is implemented on a standard PC and works in realtime.},
keywords={},
doi={},
ISSN={},
month={August},}
Copy
TY - JOUR
TI - Blind Source Separation for Moving Speech Signals Using Blockwise ICA and Residual Crosstalk Subtraction
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1941
EP - 1948
AU - Ryo MUKAI
AU - Hiroshi SAWADA
AU - Shoko ARAKI
AU - Shoji MAKINO
PY - 2004
DO -
JO - IEICE TRANSACTIONS on Fundamentals
SN -
VL - E87-A
IS - 8
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - August 2004
AB - This paper describes a real-time blind source separation (BSS) method for moving speech signals in a room. Our method employs frequency domain independent component analysis (ICA) using a blockwise batch algorithm in the first stage, and the separated signals are refined by postprocessing using crosstalk component estimation and non-stationary spectral subtraction in the second stage. The blockwise batch algorithm achieves better performance than an online algorithm when sources are fixed, and the postprocessing compensates for performance degradation caused by source movement. Experimental results using speech signals recorded in a real room show that the proposed method realizes robust real-time separation for moving sources. Our method is implemented on a standard PC and works in realtime.
ER -