Blind Separation Of Speech Signals Research Articles

A new blind separation method for under-determined speech signals based on single source interval pre-extraction

A new method is proposed to realize the blind separation of speech signals under underdetermined conditions. Before estimating the mixed parameters, the single-source interval pre-extraction operation first "filters" out a part which is obviously not a single source. The time-frequency interval of the source analysis domain category.The simulation results show the good performance of the algorithm.

Hybrid Source Prior Based Independent Vector Analysis for Blind Separation of Speech Signals

Blind Source Separation (BSS) application is a delinquent issue in a complex reverberant environment with changing room geometric dimensions and an increasing number of speech sources. The BSS application issue is determined by the independent component analysis that usually manipulates higher-order statistical approaches. However, the permutation between desired speech sources remains a challenging issue for BSS applications. The permutation problem is been rectified by Independent Vector Analysis (IVA) for BSS applications in the frequency domain. The performance dependency of the IVA approach solely relies on the selection of appropriate source-prior to preserve the inter-frequency dependencies between the same speech source amongst different frequency bins. Therefore, a hybrid model for the IVA method is presented, which comprises of multivariate generalized Gaussian and super-Gaussian distribution source priors to model low as well as high amplitudes speech signals. The weights of the hybrid model between multivariate Gaussian and generalized Gaussian are assigned in accordance to the energy of the observed non-stationary speech mixture signal. In the simulations, different speech mixtures are generated from various speech sources by simulated room model. The proposed approach evaluates the blind separation performance in terms of signal-to-distortion ratio (SDR) and is compared with well-known BSS methods. The results show an improvement of the proposed methodology for non-stationary speech signals over the state-of-the-art IVA models having a fixed source prior.

“A speech recognizer” a tool to recognize the high clarity speech signal based on existing speech using ISCA

In this research a new way introduced for solving the underdetermined blind speech signal separation problem when the number of observation is less than the sources for which the ICA is no longer applicable, which enhance the time complexity for separation of signal. To resolve that, Improved Sparse Component Analysis (ISCA) is introduced to exploit the sparse nature of TF domain, which adopt a two-step processing that contains mixing matrix estimation followed by separation of source. This ISCA is based on fuzzy c-means with Particle swarm optimization (PSO) algorithm for mixed matrix Estimation. In our work PSO is used to separate the accurate voice signal from the random mixed signal by finding the best optima solution in the cluster part. Then the source signal separation is carried out based on the shortest path. These initial processing is done and verified by Mat lab and hardware description language is generated using HDL coder and it is synthesized using Xilinx ISE. The final result illustrates that the proposed system has an improved performance in terms of SNR, Efficiency and Accuracy.

Two Model-Based EM Algorithms for Blind Source Separation in Noisy Environments

The problem of blind separation of speech signals in the presence of noise using multiple microphones is addressed. Blind estimation of the acoustic parameters and the individual source signals is carried out by applying the expectation-maximization (EM) algorithm. Two models for the speech signals are used, namely an unknown deterministic signal model and a complex-Gaussian signal model. For the two alternatives, we define a statistical model and develop EM-based algorithms to jointly estimate the acoustic parameters and the speech signals. The resulting algorithms are then compared from both theoretical and performance perspectives. In both cases, the latent data (differently defined for each alternative) are estimated in the E-step, where in the M-step, the two algorithms estimate the acoustic transfer functions of each source and the noise covariance matrix. The algorithms differ in the way the clean speech signals are used in the EM scheme. When the clean signal is assumed deterministic unknown, only the a posteriori probabilities of the presence of each source are estimated in the E-step, whereas their time–frequency coefficients are the parameters that are estimated in the M-step using the minimum variance distortionless response beamformer. If the clean speech signals are modeled as complex Gaussian signals, their power spectral densities are estimated in the E-step using the multichannel Wiener filter output. The proposed algorithms were tested using reverberant noisy mixtures of two speech sources in different reverberation and noise conditions.

A DUET-Based Method for Blind Separation of Speech Signals in Reverberant Environments

A DUET-Based Method for Blind Separation of Speech Signals in Reverberant Environments

Exploiting all combinations of microphone sensors in overdetermined frequency domain blind separation of speech signals

AbstractA new approach to overdetermined frequency domain blind source separation (BSS) of speech signals which exploits all combinations of observations and hence varying inter microphone spacings is proposed. The observations are divided into subgroups so that conventional frequency domain BSS algorithms can be used. By evaluating the separation performance obtained from each group on the basis of approximately measuring the independence of separated signals, the output of the group that has the best performance among all groups on a frequency‐by‐frequency basis is chosen as the overall output. The separated signals of the overall system are then obtained by transforming their frequency domain representations into the time domain. Simulation results based on speech signals confirm that the proposed approach has better performance based on the performance index (PI) as compared with a conventional scheme using only one microphone group and an existing overdetermined frequency domain BSS algorithm. Copyright © 2010 John Wiley & Sons, Ltd.

Blind separation of speech signals based on wavelet transform and independent component analysis

Speech signals in frequency domain were separated based on discrete wavelet transform (DWT) and independent component analysis (ICA). First, mixed speech signals were decomposed into different frequency domains by DWT and the subbands of speech signals were separated using ICA in each wavelet domain; then, the permutation and scaling problems of frequency domain blind source separation (BSS) were solved by utilizing the correlation between adjacent bins in speech signals; at last, source signals were reconstructed from single branches. Experiments were carried out with 2 sources and 6 microphones using speech signals at sampling rate of 40 kHz. The microphones were aligned with 2 sources in front of them, on the left and right. The separation of one male and one female speeches lasted 2.5 s. It is proved that the new method is better than single ICA method and the signal to noise ratio is improved by 1 dB approximately.

Searching-and-averaging method of underdetermined blind speech signal separation in time domain

Underdetermined blind signal separation (BSS) (with fewer observed mixtures than sources) is discussed. A novel searching-and-averaging method in time domain (SAMTD) is proposed. It can solve a kind of problems that are very hard to solve by using sparse representation in frequency domain. Bypassing the disadvantages of traditional clustering (e.g., K-means or potential-function clustering), the durative- sparsity of a speech signal in time domain is used. To recover the mixing matrix, our method deletes those samples, which are not in the same or inverse direction of the basis vectors. To recover the sources, an improved geometric approach to overcomplete ICA (Independent Component Analysis) is presented. Several speech signal experiments demonstrate the good performance of the proposed method.

Effect of Central Limit Theorem non-compliance on blind separation of speech by negentropy maximization

In this paper the blind separation of speech signals from their convoluted mixtures using frequency domain fixed-point independent component analysis algorithm, based on negentropy maximization, is presented. We also discuss fundamental problems of fixed-point ICA by negentropy maximization arising in the separation of the speech signal due to disobedience of the Central Limit Theorem (CLT) by the mixed speech data in the frequency domain. The experimental evidences show that CLT failure is happening due to the spectral sparseness of sources. We also present a blind method to mitigate the negative effects of this by combining null beamforming with the ICA. This combination gives a good result under the low reverberation conditions.

Blind separation of speech signals in the frequency domain

Blind source separation algorithms try to reconstruct original signals, e.g., multiple speakers, from knowledge of their superpositions, using solely the mutual statistical independence of the source signals as criterion for separation. However, application of existing algorithms to acoustic superpositions is limited by the complex nature of room transfer functions and by the use of nonlinear computations. Expanding on our previous work, we linearize the acoustic source separation problem by moving to the frequency domain [Anemüller and Gramss, DAGA (1998)] and eliminate the need for computation of nonlinear functions by using a multiple decorrelation approach. Thus, our algorithm exploits the highly redundant structure of, e.g., speech signals in order to reduce the computational cost. Results of separation experiments are presented.