Grouping Separated Frequency Components by Estimating Propagation Model Parameters in Frequency-Domain Blind Source Separation

Abstract

Abstract-This paper proposes a new formulation and optimization procedure for grouping frequency components in frequency-domain blind source separation (BSS). We adopt two separation techniques, independent component analysis (ICA) and time-frequency (T-F) masking, for the frequency-domain BSS. With ICA, grouping the frequency components corresponds to aligning the permutation ambiguity of the ICA solution in each frequency bin. With T-F masking, grouping the frequency components corresponds to classifying sensor observations in the time-frequency domain for individual sources. The grouping procedure is based on estimating anechoic propagation model parameters by analyzing ICA results or sensor observations. More specifically, the time delays of arrival and attenuations from a source to all sensors are estimated for each source. The focus of this paper includes the applicability of the proposed procedure for a situation with wide sensor spacing where spatial aliasing may occur. Experimental results show that the proposed procedure effectively separates two or three sources with several sensor configurations in a real room, as long as the room reverberation is moderately low.

Selective Listening Point Audio Based on Blind Signal Separation and Stereophonic Technology

“…Either A or B was the same signal as X, and the other was the comparison signal (Eq. (20)). However, subjects did not know which signals were reference or comparison.…”

“…The phase component of that vector contains geometrical information of the virtual sources. In [11], [16]- [20], this geometrical information was used to solve the permutation problem of FD-ICA.…”

“…Therefore, we define operation φ(·) on the transfer function vectors to extract the relative phase at each microphone [20]:…”

Stereo Source Separation and Source Counting with MAP Estimation with Dirichlet Prior Considering Spatial Aliasing Problem

“…This problem makes it difficult to classify the PDOA because the phase has the indeterminacy of modulus 2πk in high frequencies. [4] considered the spatial aliasing problem in a time-frequency mask approach, however, the number of sources N s should be known.…”

On the role of localization cues in binaural segregation of reverberant speech

“…Beamforming attempts to improve SNR of a source using directional information [3,8]. Other approaches perform a timefrequency decomposition of the mixture signals and use between channel level and time delay differences in each time-frequency (T-F) unit to estimate an output signal that originates from a particular direction [8,12,14,18]. These systems use localization information as a primary cue to achieve source segregation, and show rapid performance degradation as reverberation is added to the recordings.…”