Online blind speech extraction based on a locally quadratic kurtosis criteria and a preprocessing Automatic Gain Controller

Abstract

Abstract

Complex-Valued Independent Component Analysis for Online Blind Speech Extraction

“…The FastICA method is, by virtue of this property, denoted a fixed-point method. However, the FastICA is undefined for Gaussian-only mixtures, since yields a division-by-zero in the filter vector normalization stage in (6). These conclusions are already established properties of the FastICA method (see for instance [2], [13]).…”

“…The analysis is conducted by evaluating a second-order Taylor expansion (any term with an order higher than two is omitted) of at the point , where is a small perturbation vector to the optimal solution (here, the term "small" implies that ). The unity norm constraint (6) yields that the perturbation vector must satisfy , and thus the perturbed optimal solution is evaluated at a hyper-sphere. The second-order Taylor series expansion of around the point is (10) As long as the optimal solution as well as the perturbed optimal solution obey the unity norm constraint, the following relationship holds (from [13]): (11) This relationship yields (12) The term is always greater than or equal to 0, and the type (local maximum or minimum) of the optimal solution is therefore dependent on the sign of , i.e., the sign of the Kurtosis of the dominant source.…”

An adaptive blind beamformer with an integrated single-channel noise reduction method for robust realtime blind speech extraction

“…The approximation in [9,10] utilized the a-priori output signal Y k (n) in its control loop, whereas, in this paper, a set of noise-reduced signals Y k (n) and X k (n) are used instead, according to…”

“…The original formulation of the adaptive blind beamformer in [9,10] used two signals in its feedback control loop, the input signal vector X k (n) and an a-priori beamformer output signal…”

“…This paper investigates an approach where a single-channel noise reduction method [2,3] is integrated into the feedback control loop of a recently proposed adaptive blind beamforming technique [9,10]. The intended application is Blind Speech Extraction (BSE), where a dominant speech source (dominant in the kurtosis measure) is extracted from an observed convolutive mixture of sources [5,6,7,11].…”