Robust Principal Component Analysis Based Speaker Verification Under Additive Noise Conditions

In this paper, we present a Mixture Linear Prediction based approach for robust Gammatone Cepstral Coefficients extraction (MLPGCCs). The proposed method provides performance improvement of Automatic Speaker Verification (ASV) using i-vector and Gaussian Probabilistic Linear Discriminant Analysis GPLDA modeling under transmission channel noise. The performance of the extracted MLPGCCs was evaluated using the NIST 2008 database where a single channel microphone recorded conversational speech. The system is analyzed in the presence of different channel transmission noises such as Additive White Gaussian (AWGN) and Rayleigh fading at various Signals to Noise Ratio (SNR) levels. The evaluation results show that the MLPGCCs features are a promising way for the ASV task. Indeed, the speaker verification performance using the MLPGCCs proposed features is significantly improved compared to the conventional Gammatone Frequency Cepstral Coefficients (GFCCs) and Mel Frequency Cepstral Coefficients (MFCCs) features. For speech signals corrupted with AWGN noise at SNRs ranging from (-5 dB to 15 dB), we obtain a significant reduction of the Equal Error Rate (EER) ranging from 9.41% to 6.65% and 3.72% to 1.50%, compared with conventional MFCCs and GFCCs features respectively. In addition, when the test speech signals are corrupted with Rayleigh fading channel we achieve an EER reduction ranging from 23.63% to 7.8% and from 10.88% to 6.8% compared with conventional MFCCs and GFCCs, respectively. We also found that the combination of GFCCs and MLPGCCs gives the highest performance of speaker verification system. The best performance combination achieved is around EER from 0.43% to 0.59% and 1.92% to 3.88%.

In this paper, modified group delay (MODGD) features are used to model target speakers in the Total Variability Space (TVS) framework for speaker recognition. MODGD based features have been shown to improve speaker recognition performance owing to the ability of group delay functions to emphasise formants. The basis vectors of TVS are estimated using the PPCA algorithm while i-vectors for a speaker are extracted using the conventional technique. The estimation of the total variability space is simplified by a simple transformation of the supervectors. This results in a significant speed up in the estimation of hyperparameters of TVS as the computational complexity of PPCA algorithm is simpler compared to that of the conventaional procedure. This is important as the estimation procedure needs to handle large amounts data for estimation. The technique has already been shown to provide a speed up of 16$$\times $$×. The performance of the MODGD-based system is compared with that of the MFCC based system on the NIST SRE 2010 benchmark dataset. Two types of fusions are tested in this work--systems fused at the i-vector level and at the score level. A considerable performance improvement is observed in terms of the EER (Equal Error Rate) by employing these fusion techniques. A robust speaker recognition system with decreased development time is obtained as a result.

Background/foreground separation is one of the most fundamental tasks in computer vision, especially for video data. Robust PCA (RPCA) and its tensor extension, namely, Robust Tensor PCA (RTPCA), provide an effective framework for background/foreground separation by decomposing the data into low-rank and sparse components, which contain the background and the foreground (moving objects), respectively. However, in real-world applications, the video data is contaminated with noise. For example, in metal additive manufacturing (AM), the processed X-ray video to study melt pool dynamics is very noisy. RPCA and RTPCA are not able to separate the background, foreground, and noise simultaneously. As a result, the noise will contaminate the background or the foreground or both. There is a need to remove the noise from the background and foreground. To achieve the three components decomposition, a smooth sparse Robust Tensor Decomposition (SS-RTD) model is proposed to decompose the data into static background, smooth foreground, and noise, respectively. Specifically, the static background is modeled by the low-rank tucker decomposition, the smooth foreground (moving objects) is modeled by the spatio-temporal continuity, which is enforced by the total variation regularization, and the noise is modeled by the sparsity, which is enforced by the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\ell _{1}$ </tex-math></inline-formula> norm. An efficient algorithm based on alternating direction method of multipliers (ADMM) is implemented to solve the proposed model. Extensive experiments on both simulated and real data demonstrate that the proposed method significantly outperforms the state-of-the-art approaches for background/foreground separation in noisy cases. Note to Practitioners—This work is motivated by melt pool detection in metal additive manufacturing where the processed X-ray video from the monitoring system is very noisy. The objective is to recover the background with porosity defects and the foreground with melt pool in the presence of noise. Existing methods fail to separate the noise from the background and foreground since RPCA and RTPCA have only two components, which cannot explain the three components in the data. This paper puts forward a smooth sparse Robust Tensor Decomposition by decomposing the tensor data into low-rank, smooth, and sparse components, respectively. It is a highly effective method for background/foreground separation in noisy case. In the case studies on simulated video and X-ray data, the proposed method can handle non-additive noise, and even the case of high noise-ratio. In the proposed algorithm, there is only one tuning parameter <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda $ </tex-math></inline-formula> . Based on the case studies, our method achieves satisfying performance by taking any <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\lambda \in [{0.2,1}]$ </tex-math></inline-formula> with anisotropic total variation regularization. With this observation, practitioners can apply the proposed method without extensive parameter tuning work. Furthermore, the proposed method is also applicable to other popular industrial applications. Practitioners can use the proposed SS-RTD for degradation processes monitoring, where the degradation image contains the static background, anomaly, and random disturbance, respectively.

Sparse and low rank matrix decomposition is a method that has recently been developed for estimating different components of hyperspectral data. The rank component is capable of preserving global data structures of data, while a sparse component can select the discriminative information by preserving details. In order to take advantage of both, we present a novel decision fusion based on joint low rank and sparse component (DFJLRS) method for hyperspectral imagery in this paper. First, we analyzed the effects of different components on classification results. Then a novel method adopts a decision fusion strategy which combines a SVM classifier with the information provided by joint sparse and low rank components. With combination of the advantages, the proposed method is both representative and discriminative. The proposed algorithm is evaluated using several hyperspectral images when compared with traditional counterparts.

RPCA (Robust Principal Component Analysis) recovers sparse and low rank components from the original observation data, RPCA commonly uses ADM (Alternate Direction Method) for solving, the efficiency of algorithm depends on the nuclear norm optimization solution, that is SVD. And the application of RPCA in computer vision, image and video and so on, large amounts of data, such as images and video, make it difficult for large-scale data SVD. In this paper, we used the random matrix algorithm to improve the SVD, respectively the Count Sketch algorithm, the Prototype Randomized k-SVD algorithm and the Faster Randomized k-SVD algorithm, The main idea is to reduce the size of the original large-scale data matrix and sample randomly. Using the random projection algorithm to obtain an approximation of the original matrix, and operate QR decomposition of the approximate matrix, get the unitary matrix corresponding to the approximate matrix, and do the corresponding SVD, finally we can get the results which is similar to the original matrix calculation. Obtaining approximation of the original data matrix. But the cost time and space have been greatly optimized. Simulation experiments based on single image and video foreground detection show that the proposed methods can greatly improve the efficiency of RPCA iterative optimization.

We propose a joint registration and segmentation method for cardiac perfusion images using robust PCA (RPCA) to decompose the time series into a low rank and sparse component. Registration maximizes the smoothness of the intensity signal in the low rank component. Segmentation minimizes the sparse component's pixel intensity difference with other pixels having the same label. B-splines are used to combine the registration and segmentation costs. Tests on real patient datasets show the improved registration and segmentation accuracy over conventional methods that perform them separately.

Digital subtraction angiography (DSA) is one of imaging methods using X-ray image for clear visualization of the vessel information during intervention with a catheter. In order to obtain a fine DSA image, patients have to hold their breathing. However, steady breath hold is a burden for the patients and is sometimes difficult for elder patients. We propose a blood vessel enhancement method with consecutive digital angiographic images acquired under the natural breathing. Robust principal component analysis (RPCA) is used to enhance blood vessel information from consecutive angiographic images acquired under the natural breathing. RPCA can separate the consecutive images into a low-rank component and a sparse component. The information of contrast media is included in the sparse component. We implemented it on GPU and applied the proposed method to 13 sets of angiographic images and confirmed that it enables to generate satisfactory enhanced angiographic images.

Community detection in network-type data provides a powerful tool in analyzing and understanding real-world systems. In fact, community detection approaches aim to reduce the network&#x2019;s dimensionality and partition it into a set of disjoint clusters or communities. However, real networks are usually corrupted with noise or outliers which affect the detected community structure quality. In this paper, a new robust community detection algorithm that is capable of recovering a clean or a smoothed version of the corrupted graph and detecting the correct community structure is introduced. The proposed approach combines robust principal component analysis (RPCA) and symmetric nonnegative matrix factorization (SymNMF) in a single optimization problem. The proposed problem is solved under the framework of alternating direction methods of multipliers (ADMM). In particular, the corrupted adjacency matrix is decomposed into a low-rank and sparse components using RPCA and the community structure is detected by applying SymNMF to the extracted low-rank component. Extensive experiments that have been conducted on real and simulated binary and weighted networks show that the proposed approach significantly outperforms existing algorithms in detecting the correct community structure even in grossly corrupted networks.

Exploring kernel discriminant analysis for speaker verification with limited test data

This paper analyses the short utterance probabilistic linear discriminant analysis (PLDA) speaker verification with utterance partitioning and short utterance variance (SUV) modelling approaches. Experimental studies have found that instead of using single long-utterance as enrolment data, if long enrolled utterance is partitioned into multiple short utterances and average of short utterance i-vectors is used as enrolled data, that improves the Gaussian PLDA (GPLDA) speaker verification. This is because short utterance i-vectors have speaker, session and utterance variations, and utterance-partitioning approach compensates the utterance variation. Subsequently, SUV-PLDA is also studied with utterance partitioning approach, and utterance partitioning-based SUV-GPLDA system shows relative improvement of 9% and 16% in EER for NIST 2008 and NIST 2010 truncated 10sec-10sec evaluation condition as utterance partitioning approach compensates the utterance variation and SUV modelling approach compensates the mismatch between full-length development data and short-length evaluation data.

This paper analyses the probabilistic linear discriminant analysis (PLDA) speaker verification approach with limited development data. This paper investigates the use of the median as the central tendency of a speaker's i-vector representation, and the effectiveness of weighted discriminative techniques on the performance of state-of-the-art length-normalised Gaussian PLDA (GPLDA) speaker verification systems. The analysis within shows that the median (using a median fisher discriminator (MFD)) provides a better representation of a speaker when the number of representative i-vectors available during development is reduced, and that further, usage of the pair-wise weighting approach in weighted LDA and weighted MFD provides further improvement in limited development conditions. Best performance is obtained using a weighted MFD approach, which shows over 10% relative improvement in EER over the baseline GPLDA system on mismatched and interview-interview conditions.

Chemical exchange saturation transfer (CEST) MRI requires the acquisition of multiple saturation-weighted images and can last several minutes. Misalignments among these images, which are often due to the inevitable motion of the subject, will corrupt CEST contrast maps and result in large quantification errors. Therefore, the registration of the CEST series is critical. However, registration is challenging since common intensity-based registration algorithms may fail to differentiate CEST signals from motion artifacts. Herein, we studied how different patterns of motion affect CEST quantification and proposed a cascaded two-step registration scheme by utilizing features extracted from the entire Z-spectral image series instead of direct registration to a single image. The proposed approach is conducted in two stages: during the first coarse registration, the Z-spectral image series is decomposed by robust principal component analysis (RPCA) to separate CEST contrast from motion. The recomposed image series using only the low-rank component, which contains minimized motion, are averaged to generate a reference for the alignment of the image series. To further remove residual misalignments, the coarse registration is followed by a refinement stage, which uses PCA iteratively to generate motionless synthetic reference series with the first few principal components (PCs) that correspond to CEST contrast. In the end, the quality check is performed to exclude the images with unsuccessful registration. The proposed registration scheme (RPCA + PCA_R) was assessed by both phantom experiments and in vivo data of tumor-bearing mouse brain, with simulated random rigid motion in different patterns applied to the acquired static Z-spectral image series. For comparison, previous correction schemes using an explicit image [either S0 or Ssat(∆ω)] as registration reference were also performed, named as S0_R and Ssat_R respectively. To illustrate the advantage of combination of RPCA and PCA, registration was also exploited using either only the RPCA-based method (RPCA_R) or only the PCA-based one (PCA_R). Compared with the above four methods, RPCA + PCA_R allowed for more accurate correction of the corrupted Z-spectral images, exhibiting smaller MTRasym(∆ω) error maps and lower residual Z-spectra referring to the static data. Among all the five correction methods, the corrected Z-spectral image series by RPCA + PCA_R and the resulting MTRasym(∆ω) maps achieved the highest correlation coefficients (CC) with respect to the static ones. The registration scheme of RPCA + PCA_R provides robust motion correction between two specific Z-spectral images and among an entire image series, through extraction of the static component from the entire Z-spectra set and inclusion of a PCA-based refinement step. Therefore, this method can help improve CEST acquisition and quantification.

Narrow-band interference (NBI) is a critical issue for synthetic aperture radar (SAR), in which the imaging quality can be degraded severely. To suppress NBI effectively, a novel interference suppression algorithm using robust principal component analysis (RPCA) based signal separation in time-frequency domain is proposed. The RPCA algorithm is introduced for signal separation in the time-frequency domain for the first time. The fundamental assumption of RPCA is that a matrix can be modeled as a combination of a low-rank matrix and a sparse counterpart. In terms of the SAR echo, the short time Fourier transformation (STFT) matrix of mixed signals (i.e., useful SAR signals and NBIs) well fits the assumption of RPCA. Based on this property, radar echoes are first transformed into the time-frequency domain by STFT to form an STFT matrix. Then, the RPCA algorithm is used to decompose the STFT matrix into a low-rank matrix (i.e., NBIs) and a sparse matrix (i.e., useful signals). Finally, the NBIs can be reconstructed and subtracted from the echoes to realize the interference suppression. The experimental results of simulated and measured data demonstrate that the proposed algorithm not only can suppress interference effectively, but also preserve the useful information as much as possible.

This paper presents Program Guardian, which is a speaker recognition-based screening system for smart TV. The system identifies a specific person from his or her voice such that the smart TV can provide suitable programs for that person. This system is based on a robust speaker recognition system that uses robust principal component analysis (RPCA) and a sparse representation classifier (SRC). First, i-vectors that are generated from supervectors of Gaussian mixture models (GMMs) are used to generate the basic atoms of an over-complete dictionary. The i-vectors are then transformed using RPCA. The SRC is produced from transformed i-vector-based RPCA vectors. Finally, the sparse representation classifier corresponding to the target speaker with the least reconstruction error is constructed. NIST speaker recognition evaluation data base is used in our experiment. The results show that the proposed speaker recognition system is feasible and offers advantages over accuracy.

The mismatch between the training conditions and the test conditions severely degrades the performance of speaker verification. Aiming at solving this problem, this paper presents a method which uses a combination of improved nonnegative matrix factorization (IMNMF) and mel frequency cepstral coefficients (MFCC) with feature warping for improving identity-vector (i-vector) speaker verification performance in noisy environment. Unlike the traditional nonnegative matrix factorization (NMF), IMNMF uses extra free basis vectors to capture the features which are not included in training data, and linear constraints on dictionary atoms. Feature warping is used to remove channel noises. Therefore, the proposed method can reduce distortion of reconstructed speech while enhancing the recovered speech quality. The performance of i-vector speaker verification is evaluated by using the short utterance database and the NOISEX-92 database. The experiment results indicate that the score level fusion of feature-warped IMNMF-MFCC and feature-warped MFCC is superior to the baseline system at the equal error rate under the majority of signal-to-noise ratios (SNRs).