A Region-Growing Permutation Alignment Approach in Frequency-Domain Blind Source Separation of Speech Mixtures

• Published in: IEEE transactions on audio, speech, and language processing Vol. 19; no. 3; pp. 549 - 557
• Main Authors: Wang, Lin, Ding, Heping, Yin, Fuliang
• Format: Journal Article
• Language: English
• Published: Piscataway, NJ IEEE 01.03.2011; Institute of Electrical and Electronics Engineers
• Subjects: Acoustic signal processing; Acoustical engineering; Alignment; Applied sciences; Bins; Blind source separation; Blind source separation (BSS); Blinds; Computational efficiency; convolutive mixture; Councils; Detection, estimation, filtering, equalization, prediction; Educational programs; Exact sciences and technology; frequency domain; Frequency domain analysis; Information, signal and communications theory; Miscellaneous; Permission; permutation problem; Permutations; power ratio; region growing; Separation; Signal and communications theory; Signal processing; Signal resolution; Signal, noise; Source separation; Speech; Spreading; Telecommunications and information theory; Performance evaluation; permutation problem; Frequency dependence; Source separation; Signal estimation; Frequency domain method; Blind source separation; Frequency measurement; Blind source separation (BSS); frequency domain; power ratio; Frequency band; Signal processing; region growing; Alignment defect; convolutive mixture
• ISSN: 1558-7916; 1558-7924
• DOI: 10.1109/TASL.2010.2052244

The convolutive blind source separation (BSS) problem can be solved efficiently in the frequency domain, where instantaneous BSS is performed separately in each frequency bin. However, the permutation ambiguity in each frequency bin should be resolved so that the separated frequency components from the same source are grouped together. To solve the permutation problem, this paper presents a new alignment method based on an inter-frequency dependence measure: the powers of separated signals. Bin-wise permutation alignment is applied first across all frequency bins, using the correlation of separated signal powers; then the full frequency band is partitioned into small regions based on the bin-wise permutation alignment result. Finally, region-wise permutation alignment is performed in a region-growing manner. The region-wise permutation correction scheme minimizes the spreading of the misalignment at isolated frequency bins to others, hence to improve permutation alignment. Experiment results in simulated and real environments verify the effectiveness of the proposed method. Analysis demonstrates that the proposed frequency-domain BSS method is computationally efficient.