Search

Previous research has found that human voice can provide reliable information to be used for gender identification with a high level of accuracy. In social psychology, perceived masculinity and femininity (masculinity and femininity rated by humans) has often been considered an important feature when investigating the influence of vocal features on social behaviours. While previous studies have characterized the acoustic features that contributed to perceivers' judgements of speakers' masculinity or femininity, there is limited research on developing a machine masculinity/femininity scoring model and characterizing the independent acoustic factors that contribute to perceivers' masculinity and femininity judgements. In this work, we first propose a machine scoring model of perceived masculinity/femininity based on the Extreme Random Forest and then characterize the independent and meaningful acoustic factors that contribute to perceivers' judgements by using a correlation matrix based hierarchical clustering method. Our results show that the machine ratings of masculinity and femininity strongly correlated with the human ratings of masculinity and femininity when we used an optimal speech duration of 7 s, with a correlation coefficient of up to.63 for females and.77 for males. Nine independent clusters of acoustic measures were generated from our modelling of femininity judgements for female voices and eight clusters were found for masculinity judgements for male voices. The results revealed that, for both genders, the F0 mean is the most important acoustic measure affecting the judgement of acoustic-related masculinity and femininity. The F3 mean, F4 mean and VTL estimators were found to be highly inter-correlated and appeared in the same cluster, forming the second most significant factor in influencing the assessment of acoustic-related masculinity and femininity. Next, F1 mean, F2 mean and F0 standard deviation are independent factors that share similar importance. The voice perturbation measures, including HNR, jitter and shimmer, are of lesser importance in influencing masculinity/femininity judgements. • We modelled femininity/masculinity ratings for 129 female/96 male voices. • Extreme Random Forest regression and correlated feature clustering were used. • F0 mean accounted for 23.80% of variance for femininity and 43.5% for masculinity. • A F3 mean/F4 mean/VTL estimators was also a prominent predictor. • Differences in feature weights for femininity and masculinity ratings are discussed. [ABSTRACT FROM AUTHOR]

Spoofing attacks posed by generating artificial speech can severely degrade the performance of a speaker verification system. Recently, many anti-spoofing countermeasures have been proposed for detecting varying types of attacks from synthetic speech to replay presentations. While there are numerous effective defenses reported on standalone anti-spoofing solutions, the integration for speaker verification and spoofing detection systems has obvious benefits. In this paper, we propose a spoofing-robust automatic speaker verification system for diverse attacks based on a multi-task learning architecture. This deep learning-based model is jointly trained with time-frequency representations from utterances to provide recognition decisions for both tasks simultaneously. Compared with other state-of-the-art systems on the ASVspoof 2017 and 2019 corpora, a substantial improvement of the combined system under different spoofing conditions can be obtained. [ABSTRACT FROM AUTHOR]

Recently, neural network-based deep learning methods have been popularly applied to computer vision, speech signal processing and other pattern recognition areas. Remarkable success has been demonstrated by using the deep learning approaches. The purpose of this article is to provide a comprehensive survey for the neural network-based deep learning approaches on acoustic event detection. Different deep learning-based acoustic event detection approaches are investigated with an emphasis on both strongly labeled and weakly labeled acoustic event detection systems. This paper also discusses how deep learning methods benefit the acoustic event detection task and the potential issues that need to be addressed for prospective real-world scenarios. [ABSTRACT FROM AUTHOR]

Highlights • Multiscale fractal analysis of architectural distortion was applied. • Bidimensional EMD was applied to mammographic ROI. • Multiscale fractal dimension was applied. • Multiscale fractal properties of architectural distortions are distinct. • Results obtained are promising. Abstract Achieving a high performance for the detection and characterization of architectural distortion in screening mammograms is important for an efficient breast cancer early detection. Viewing a mammogram image as a rough surface that can be described using the fractal theory is a well-recognized approach. This paper presents a new fractal-based computer-aided detection (CAD) algorithm for characterizing various breast tissues in screening mammograms with a particular focus on distinguishing between architectural distortion and normal breast parenchyma. The proposed approach is based on two underlying assumptions: (i) monitoring the variation pattern of fractal dimension, with the changes of the image resolution, is a useful tool to distinguish textural patterns of breast tissue, (ii) the bidimensional empirical mode decomposition (BEMD) algorithm appropriately generates a multiresolution representation of the mammogram. The proposed CAD has been tested using different validation datasets of mammographic regions of interest (ROIs) extracted from the Digital Database for Screening Mammography (DDSM) database. The validation ROI datasets contain architectural distortion (AD), normal breast tissue, and AD surrounding tissue. The highest classification performance, in terms of area under the receiver operating characteristic curve, of Az = 0.95 was achieved when the proposed approach applied to distinguish 187 architectural distortion depicting regions from 2191 normal breast parenchyma regions. The obtained results validate the underlying hypothesis and demonstrate that effectiveness of capturing the variation of the fractal dimension measurements within an appropriate multiscale representation of the digital mammogram. Results also reveal that this tool has the potential of prescreening other key and common mammographic signs of early breast cancer. [ABSTRACT FROM AUTHOR]

The variety of event categories and event boundary information have resulted in limited success for acoustic event detection systems. To deal with this, we propose to utilize the long contextual information, low-dimensional discriminant global bottleneck features and category-specific bottleneck features. By concatenating several adjacent frames together, the use of contextual information makes it easier to cope with acoustic signals with long duration. Global and category-specific bottleneck features can extract the prior knowledge of the event category and boundary, which is ideally matched by the task of an event detection system. Evaluations on the UPC-TALP and ITC-IRST databases of highly variable acoustic events demonstrate the effectiveness of the proposed approaches by achieving a 5.30% and 4.44% absolute error rate improvement respectively compared to the state of art technique. [ABSTRACT FROM AUTHOR]

Although stereo information has been extensively used in computer vision tasks recently, the incorporation of stereo visual information in Audio-Visual Speech Recognition (AVSR) systems and whether it can boost the speech accuracy still remains a largely undeveloped area. This paper addresses three fundamental issues in this area: 1) Will the stereo features benefit visual and audio-visual speech recognition? 2) If so, how much information is embedded in stereo features? 3) How to encode both planar and stereo information in a compact feature vector? In this study, we propose a comprehensive study on the characteristics of both planar and stereo visual features, and extensively analyse why the stereo information can boost the visual speech recognition. Based on the different information embedded in planar and stereo features, we present a new Cascade Hybrid Appearance Visual Feature (CHAVF) extraction scheme which successfully combines planar and stereo visual information into a compact feature vector, and evaluate this novel feature on visual and audio-visual connected digit recognition and isolated phrase recognition. The results show that stereo information is capable of significantly boosting the speech recognition, and the performance of our proposed visual feature outperforms the other commonly used appearance-based visual features on both the visual and audio-visual speech recognition tasks. Particularly, our proposed planar-stereo visual feature yields approximately 21% relative improvement over the planar visual feature. To the best of our knowledge, this is the first paper that extensively evaluates the different characteristics of planar and stereo visual features, and we first show that using the stereo feature along with the planar feature can significantly boost the accuracy on a large-scale audio-visual data corpus. [ABSTRACT FROM AUTHOR]

Missing data approaches have recently been applied to speech recognition tasks to increase noise robustness. The drawback of missing data techniques is the vulnerability of the recognizer to errors in the reliability mask. This work proposes a novel group selection algorithm to perform top-down refinement of initial bottom-up reliability mask estimates with the goal of removing these errors. A novel probabilistic decision process based on normalized likelihood distances is proposed and used to evaluate the quality of a reliability mask without any a priori noise knowledge. Experimental results on a speaker identification task illustrate the ability of the combined bottom-up top-down system to significantly outperform traditional bottom-up only missing data techniques for various types of mask corruption. [ABSTRACT FROM AUTHOR]

Abstract: In this paper we address the problem of robust face recognition by formulating the pattern recognition task as a problem of robust estimation. Using a fundamental concept that in general, patterns from a single object class lie on a linear subspace (Barsi and Jacobs, 2003 ), we develop a linear model representing a probe image as a linear combination of class specific galleries. In the presence of noise, the well-conditioned inverse problem is solved using the robust Huber estimation and the decision is ruled in favor of the class with the minimum reconstruction error. The proposed Robust Linear Regression Classification (RLRC) algorithm is extensively evaluated for two important cases of robustness i.e. illumination variations and random pixel corruption. Illumination invariant face recognition is demonstrated on three standard databases under exemplary evaluation protocols reported in the literature. Comprehensive comparative analysis with the state-of-art illumination tolerant approaches indicates a comparable performance index for the proposed RLRC algorithm. The efficiency of the proposed approach in the presence of severe random noise is validated under several exemplary noise models such as dead-pixel problem, salt and pepper noise, speckle noise and Additive White Gaussian Noise (AWGN). The RLRC algorithm is found to be favorable compared with the benchmark generative approaches. [Copyright &y& Elsevier]

Abstract: Time–frequency masking has evolved as a powerful tool for tackling blind source separation problems. In previous work, mask estimation was performed with the help of well-known standard cluster algorithms. Spatial observation vectors, extracted from a set of microphones, were grouped into separate clusters, each representing a particular source. However, most off-the-shelf clustering methods are not very robust to outliers or noise in the data. This lack of robustness often leads to incorrect localization and partitioning results, particularly for reverberant speech mixtures. To address this issue, we investigate the use of observation weights and context information as means to improve the clustering performance under reverberant conditions. While the observation weights improve the localization accuracy by ignoring noisy observations, context information smoothes the cluster membership levels by exploiting the highly structured nature of speech signals in the time–frequency domain. In a number of experiments, we demonstrate the superiority of the proposed method over conventional fuzzy clustering, both in terms of localization accuracy as well as speech separation performance. [Copyright &y& Elsevier]

The performances of two perceptual properties of the peripheral auditory system, synaptic adaptation and two-tone suppression, are compared for automatic speech recognition (ASR) in an additive noise environment. A simple method of synaptic adaptation as determined by psychoacoustic observations was implemented with temporal processing of speech utilizing a zero-crossing auditory model as a pre-processing front end. The concept is similar to RASTA processing, but instead of bandpass filters, a high-pass infinite impulse response (IIR) filter is used. It is shown that rapid synaptic adaptation may be implemented by temporal processing using the zero-crossing algorithm, not otherwise implementable in the spectral domain implementation. The two-tone suppression was implemented in the zero-crossing auditory model using a companding strategy. Recognition performances with the two perceptual features were evaluated on isolated digits (TIDIGITS) corpus using continuous density HMM recognizer in white, factory, babble and Volvo noise. It is observed that synaptic adaptation performs better in stationary white Gaussian noise. In presence of non-stationary non-Gaussian noise, however, no improvements or a degradation is observed. Moreover, a reciprocal effect is observed with two-tone suppression, with better performance in non-Gaussian real-world noise and degradation in stationary white Gaussian noise. [Copyright &y& Elsevier]

Abstract: In this paper, we present a state-space formulation of a neural-network-based hidden dynamic model of speech whose parameters are trained using an approximate EM algorithm. This efficient and effective training makes use of the output of an off-the-shelf formant tracker (for the vowel segments of the speech signal), in addition to the Mel-cepstral observations, to simplify the complex sufficient statistics that would be required in the exact EM algorithm. The trained model, consisting of the state equation for the target-directed vocal tract resonance (VTR) dynamics on all classes of speech sounds (including consonant closure and constriction) and the observation equation for mapping from the VTR to Mel-cepstral acoustic measurement, is then used to recover the unobserved VTR based on the extended Kalman filter. The results demonstrate accurate estimation of the VTR, especially during rapid consonant–vowel or vowel–consonant transitions and during consonant closure when the acoustic measurement alone provides weak or no information to infer the VTR values. The practical significance of correctly identifying the VTRs during consonantal closure or constriction is that they provide target frequency values for the VTR or formant transitions from adjacent sounds. Without such target values, the VTR transitions from vowel to consonant or from consonant to vowel are often very difficult to extract accurately by the previous formant tracking techniques. With the use of the new technique reported in this paper, the consonantal VTRs and the related transitions become more reliably identified from the speech signal. [Copyright &y& Elsevier]

In this paper, we present a new approach to joint state and parameter estimation for a target-directed, nonlinear dynamic system model with switching states. The model, which was recently proposed for representing speech dynamics, is also called the hidden dynamic model (HDM). The model parameters subject to statistical estimation consist of the target vector and the system matrix (also called the "time-constants"), as well as the parameters characterizing the nonlinear mapping from the hidden state to the observation. These latter parameters are implemented in the current work as the weights of a three-layer feedforward multilayer perceptron (MLP) network. The new estimation approach presented in this paper is based on the extended Kaiman filter (EKF), and its performance is compared with the more traditional approach based on the expectation-maximization (EM) algorithm. Extensive simulation experiment results are presented using the proposed EKF-based and the EM algorithms and under the typical conditions for employing the HDM for speech modeling. The results demonstrate superior convergence performance of the EKF-based algorithm compared with the EM algorithm, but the former suffers from excessive computational loads when adopted for training the MLP weights. In all cases, the simulation results show that the simulated model output converges to the given observation sequence. However, only in the case where the MLP weights or the target vector are assumed known do the time-constant parameters converge to their true values. We also show that the MLP weights never converge to their true values, thus demonstrating the many-to-one mapping property of the feedforward MLP. We conclude from these simulation experiments that for the system to be identifiable, restrictions on the parameter space are needed. [ABSTRACT FROM AUTHOR]

• Novel anti-spoofing countermeasure for replay attacks • Speech features are generated by GANs with conditions • A CNN based post selection process is proposed • 3.26% and 8.05% EERs on the development and evaluation subsets of the ASVspoof 2017 V2.0 corpus are achieved Automatic Speaker Verification (ASV) systems have been widely applied for speaker authentication for biometric security especially in e-business scenarios. However, vulnerabilities of the ASV technology have been discovered and have generated much interest to design anti-spoofing countermeasures. Serious threats can be posed by replay attacks, which are difficult to detect and easy to mount with accessible devices. In this paper, an efficient replay anti-spoofing countermeasure based on data augmentation with post selection is proposed. The auxiliary classifier generative adversarial network (AC-GAN) is adopted to generate more speech samples with diverse variants. To select generated samples of high quality and avoid the bias caused by human subjective perception, we also propose a convolutional neural network (CNN) based post-filter. By integrating data augmentation and post selection approaches, issues of over-fitting and lack of generalization can be significantly alleviated with extra informative training data. The proposed anti-spoofing countermeasure is evaluated on the ASVspoof 2017 Version 2.0 database. Experimental results measured by equal error rates (EERs) indicate a promising improvement over the development and evaluation subsets. This provides the motivation for novel audio data augmentation and also promotes the future research on generation selection in the application of speaker spoofing detection. [ABSTRACT FROM AUTHOR]

Highlights from the article: Deep learning employs deep neural networks (DNNs), which are neural networks with more than one hidden layer, with recently developed initialization and training strategies using massive amounts of diverse data as examples. The paper provides an overview of recent approaches to deep learning as applied to a range of speech processing tasks, primarily for automatic speech recognition (ASR), but also text-to-speech and speaker, language, and emotion recognition. The authors propose integrating deep neural network (DNN)-HMM technique with the HiLAM method where the state alignment information from the HiLAM is used to discriminatively train a DNN to further improve the system performance.

In the recent past years, deep-learning-based machine learning methods have demonstrated remarkable success for a wide range of learning tasks in multiple domains. They are suitable for complex classification and regression problems in applications such as computer vision, speech recognition and other pattern analysis branches. The purpose of this article is to contribute a timely review and introduction of state-of-the-art and popular discriminative DNN, CNN and RNN deep learning techniques, the basic framework and algorithms, hardware implementations, applications in speech, and the overall benefits of deep learning. [ABSTRACT FROM AUTHOR]

• A novel concept and framework is introduced to directly match 2D and 3D features. • The framework is used for 6-DOF pose estimation, and camera and visual localization. • Current techniques usually rely on Structure from Motion models for localization. • Proposed framework can be used for any type of 3D models and point clouds. • The method outperforms state-of-the-art techniques on benchmark datasets. This paper presents a novel approach for cross-domain descriptor matching between 2D and 3D modalities. The 2D-3D matching is applied to localize 2D images in 3D point clouds. Direct cross-domain matching allows our technique to localize images in any type of 3D point cloud without any constraints on the nature or mechanism by which it is obtained. We propose a learning based framework, called Desc-Matcher, to directly match features between the two modalities. A dataset of 2D and 3D features with corresponding locations in images and point clouds is generated to train the Desc-Matcher. To estimate the pose of an image in any 3D cloud, keypoints and feature descriptors are extracted from the query image and the point cloud. The trained Desc-Matcher is then used to match the features from the image and the point cloud. A robust pose estimator is used to predict the location and orientation of the query image from the corresponding positions of the matched 2D and 3D features. We carried out an extensive evaluation of the proposed method for indoor and outdoor scenarios and with different types of point clouds to verify the feasibility of our approach. Experimental results show that the proposed approach can reliably estimate the 6-DOF poses of query cameras in any type of 3D point cloud with high precision. We achieved average median errors of 1.09 c m / 0. 27 ∘ and 19 c m / 0. 39 ∘ on the Stanford and Cambridge datasets, respectively. [ABSTRACT FROM AUTHOR]

In this paper, we propose a consolidated framework for the automatic selection of the most discriminant subbands for the problem of face recognition. Essentially, the face images are transformed into textures using the linear binary pattern (LBP) approach, these texturized-faces undergo the wavelet packet decomposition resulting in several subband images. We propose to use the energy features to effectively represent these subband images. The underlying statistical patterns of the data are harnessed in form of information-theoretic metrics to select the most discriminant subbands. The proposed algorithms are extensively evaluated on several standard databases and are shown to always pick the most significant subbands resulting in better performance. The proposed algorithms are entirely generic and do not depend on the selection of features or/and classifiers. [ABSTRACT FROM AUTHOR]

Bounded Component Analysis (BCA) solves the Blind Source Separation (BSS) problem based on geometric assumptions. This paper introduces a new proof of a BCA contrast function, derived from elementary matrices, Gauss–Jordan elimination and convex geometry. The new proof and further analysis provide additional insight into a key assumption of BCA. In addition, an interpretation is presented to clarify one of the limitations of the instantaneous BCA algorithm. Experiments on audio sources support our analysis. [ABSTRACT FROM AUTHOR]

In this research, the concept of class-specific dictionaries is proposed for iris recognition. Essentially, the query image is represented as a linear combination of training images from each class. The well-conditioned inverse problem is solved using least squares regression and the decision is ruled in favor of the class with the most precise estimation. An enhanced modular approach is further proposed to counter noise due to imperfect segmentation of the iris region. As such iris images are partitioned and individual decisions of all sectors are fused using an efficient fusion algorithm. The proposed algorithm is compared to the state-of-the-art Sparse Representation Classification (SRC) with Bayesian fusion for multiple sectors. The proposed approach has shown to comprehensively outperform the SRC algorithm on standard databases. Complexity analysis of the proposed algorithm shows decisive superiority of the proposed approach. [ABSTRACT FROM AUTHOR]

One highlight of deep networks is their ability to automatically learn useful representations from raw inputs. Hence, domain knowledge, generic priors and feature extraction are no longer needed. It is however still a challenging task to train deep networks, because they require extensive human expertise to choose the type of model and its parameters to ensure that the system is properly trained. In this work, we propose a novel feature learning framework based on the marginalised stacked auto-encoder which does not need practitioners to have any deep learning specific knowledge. We applied this method on visual speech recognition, and the performance of our proposed method outperforms the other feature extraction methods with a 2% improvement in the accuracy for speaker independent systems. This method is also a universal solution which can be used for any deep learning based tasks. Therefore, we also verified our method on a popular hand written digit recognition database MNIST, and experimental results showed that our proposed method with an error rate of 1.30% is comparable to the best models tuned by experts. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel adaptive kernel for the radial basis function neural networks. The proposed kernel adaptively fuses the Euclidean and cosine distance measures to exploit the reciprocating properties of the two. The proposed framework dynamically adapts the weights of the participating kernels using the gradient descent method, thereby alleviating the need for predetermined weights. The proposed method is shown to outperform the manual fusion of the kernels on three major problems of estimation, namely nonlinear system identification, patter classification and function approximation. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel approach of face identification by formulating the pattern recognition problem in terms of linear regression. Using a fundamental concept that patterns from a single-object class lie on a linear subspace, we develop a linear model representing a probe image as a linear combination of class-specific galleries. The inverse problem is solved using the least-squares method and the decision is ruled in favor of the class with the minimum reconstruction error. The proposed Linear Regression Classification (LRC) algorithm falls in the category of nearest subspace classification. The algorithm is extensively evaluated on several standard databases under a number of exemplary evaluation protocols reported in the face recognition literature. A comparative study with state-of-the-art algorithms clearly reflects the efficacy of the proposed approach. For the problem of contiguous occlusion, we propose a Modular LRC approach, introducing a novel Distance-based Evidence Fusion (DEF) algorithm. The proposed methodology achieves the best results ever reported for the challenging problem of scarf occlusion. [ABSTRACT FROM PUBLISHER]

This paper presents a confidence-based late fusion framework and its application to audio-visual biometric identification. We assign each biometric matcher a confidence value calculated from the matching scores it produces. Then a transformation of the matching scores is performed using a novel confidence-ratio ( C -ratio) i.e., the ratio of a matcher confidence obtained at the test phase to the corresponding matcher confidence obtained at the training phase. We also propose modifications to the highest rank and Borda count rank fusion rules to incorporate the matcher confidence. We demonstrate by experiments that our proposed confidence-based fusion framework is more robust compared to the state-of-the-art late (score- and rank-level) fusion approaches. [ABSTRACT FROM AUTHOR]

• Comparative analysis of data scaling parameters, input feature selection with univariate/multivariate preferences. • Evaluation of the Long Short Term Memory model using Data Segmentation structures, varied time-horizons, and case studies. • An ensemble boosting approach with an adjustable weighting mechanism is developed to optimize forecast accuracy. • Runtime assessment and viability of the proposed approach. Photovoltaics (PVs) hold the promise of sustainable electricity production. However, PV output is significantly influenced by variations in terrestrial solar radiation and other weather factors, causing problems in unit commitment, economic power dispatch and reliable electricity distribution from grid-connected, hybrid and off-grid PV plants. Thus, accurate and consistent PV output prediction is crucial for utilities companies and researchers. Current initiatives enlist a plethora of forecasting techniques, including: statistical models, physical models, artificial neural networks and deep learning algorithms. Nevertheless, most of these approaches suffer from computational costs, complex models and uncertainties. Hence, this research employed an ensemble-based Long Short-Term Memory (LSTM) algorithm comprising 10 component LSTM models. The method compares the effects of different data segmentations (three-months to one-day) and is based on varying time-horizons (14-days to 5-mins) in order to compare the effects of seasonal and periodic variations on time-series data and PV output forecast. The comparison is then used to minimise uncertainty by implementing grid search technique. Subsequently, the model's performance was evaluated using MAPE analysis for two cases involving erratic weather conditions and PV output in two specific days of the year 2020. From the evaluation, the best prediction was found to be for the two-week dataset with MAPE of 6.02. This approach combined with online data acquisition from the Yulara Solar System power plant in central Australia leads to a more practical, computationally economical and robust PV power generation forecasting technique. [ABSTRACT FROM AUTHOR]

Abstract: Geoscientific data interpretation is a challenging task, which requires the detection and synthesis of complex patterns within data. As a first step towards better understanding this interpretation process, our research focuses on quantitative monitoring of interpreters' brain responses associated with geoscientific target spotting. This paper presents a method that profiles brain responses using electroencephalography (EEG) to detect P300-like responses that are associated with target spotting for complex geoscientific data. In our experiment, eight interpreters with varying levels of expertise and experience were asked to detect features, which are likely to be copper–gold rich porphyry systems within magnetic geophysical data. The target features appear in noisy background and often have incomplete shape. Magnetic images with targets and without targets were shown to participants using the “oddball” paradigm. Event related potentials were obtained by averaging the EEG epochs across multiple trials and the results show delayed P3 response to the targets, likely due to the complexity of the task. EEG epochs were classified and the results show reliable single trial classification of EEG responses with an average accuracy of 83%. The result demonstrated the usability of the P300-like responses to quantify the geoscientific target spotting performances. [Copyright &y& Elsevier]

Abstract: This paper attempts to overcome the local convergence problem of the Expectation Maximization (EM) based training of the Hidden Markov Model (HMM) in speech recognition. We propose a hybrid algorithm, Simulated Annealing Stochastic version of EM (SASEM), combining Simulated Annealing with EM that reformulates the HMM estimation process using a stochastic step between the EM steps and the SA. The stochastic processes of SASEM inside EM can prevent EM from converging to a local maximum and find improved estimation for HMM using the global convergence properties of SA. Experiments on the TIMIT speech corpus show that SASEM obtains higher recognition accuracies than the EM. [Copyright &y& Elsevier]

This paper attempts to overcome the tendency of the expectation-maximization (EM) algorithm to locate a local rather than global maximum when applied to estimate the hidden Markov model (HMM) parameters in speech signal modeling. We propose a hybrid algorithm for estimation of the HMM in automatic speech recognition (ASR) using a constraint-based evolutionary algorithm (EA) and EM, the CEL-EM. The novelty of our hybrid algorithm (CEL-EM) is that it is applicable for estimation of the constraint-based models with many constraints and large numbers of parameters (which use EM) like HMM. Two constraint-based versions of the CEL-EM with different fusion strategies have been proposed using a constraint-based EA and the EM for better estimation of HMM in ASR. The first one uses a traditional constraint-handling mechanism of EA. The other version transforms a constrained optimization problem into an unconstrained problem using Lagrange multipliers. Fusion strategies for the CEL-EM use a staged-fusion approach where EM has been plugged with the EA periodically after the execution of EA for a specific period of time to maintain the global sampling capabilities of EA in the hybrid algorithm. A variable initialization approach (VIA) has been proposed using a variable segmentation to provide a better initialization for EA in the CEL-EM. Experimental results on the TIMIT speech corpus show that CEL-EM obtains higher recognition accuracies than the traditional EM algorithm as well as a top-standard EM (VIA-EM, constructed by applying the VIA to EM). [ABSTRACT FROM AUTHOR]

In this work, a new class of stochastic gradient algorithm is developed based on q-calculus. Unlike the existing q-LMS algorithm, the proposed approach fully utilizes the concept of q-calculus by incorporating a time-varying q parameter. The proposed enhanced q-LMS (Eq-LMS) algorithm utilizes a novel, parameterless concept of error-correlation energy and normalization of signal to ensure high convergence, stability and low steady-state error. The proposed algorithm automatically adapts the learning rate with respect to the error. For evaluation purposes the system identification problem is considered. The necessary condition of convergence for the proposed algorithm is analyzed, and the validation of analytical findings and simulation results is discussed. Extensive experiments show better performance of the proposed Eq-LMS algorithm compared to the standard q-LMS approach. [ABSTRACT FROM AUTHOR]

In this paper, we propose a novel approach for 3-D rain field reconstruction using satellite signals. It uses the estimated signal-to-noise ratio (SNR) at the ground receivers for low-earth orbit (LEO) satellites and, thus, indirectly estimates the path-integrated rain attenuation of the microwave communication links. A least-squares algorithm is employed to perform the 3-D tomographic reconstruction of the rain field. The proposed system model consists of an LEO satellite with a realistic overpass trajectory and multiple ground receivers with SNR estimators. Two synthetic rain events near the Great Barrier Reef in Australia are used to test the reconstruction outcome. Simulation results suggest that the reconstructed rain field has close agreement with the synthetic rain field. [ABSTRACT FROM AUTHOR]

In this research, we propose a novel fractional gradient descent-based learning algorithm (FGD) for the radial basis function neural networks (RBF-NN). The proposed FGD is the convex combination of the conventional, and the modified Riemann-Liouville derivative-based fractional gradient descent methods. The proposed FGD method is analyzed for an optimal solution in a system identification problem, and a closed form Wiener solution of a least square problem is obtained. Using the FGD, the weight update rule for the proposed fractional RBF-NN (FRBF-NN) is derived. The proposed FRBF-NN method is shown to outperform the conventional RBF-NN on four major problems of estimation namely nonlinear system identification, pattern classification, time series prediction and function approximation. [ABSTRACT FROM AUTHOR]

We present a framework to automatically detect and remove shadows in real world scenes from a single image. Previous works on shadow detection put a lot of effort in designing shadow variant and invariant hand-crafted features. In contrast, our framework automatically learns the most relevant features in a supervised manner using multiple convolutional deep neural networks (ConvNets). The features are learned at the super-pixel level and along the dominant boundaries in the image. The predicted posteriors based on the learned features are fed to a conditional random field model to generate smooth shadow masks. Using the detected shadow masks, we propose a Bayesian formulation to accurately extract shadow matte and subsequently remove shadows. The Bayesian formulation is based on a novel model which accurately models the shadow generation process in the umbra and penumbra regions. The model parameters are efficiently estimated using an iterative optimization procedure. Our proposed framework consistently performed better than the state-of-the-art on all major shadow databases collected under a variety of conditions. [ABSTRACT FROM PUBLISHER]

Inexpensive structured light sensors can capture rich information from indoor scenes, and scene labeling problems provide a compelling opportunity to make use of this information. In this paper we present a novel conditional random field (CRF) model to effectively utilize depth information for semantic labeling of indoor scenes. At the core of the model, we propose a novel and efficient plane detection algorithm which is robust to erroneous depth maps. Our CRF formulation defines local, pairwise and higher order interactions between image pixels. At the local level, we propose a novel scheme to combine energies derived from appearance, depth and geometry-based cues. The proposed local energy also encodes the location of each object class by considering the approximate geometry of a scene. For the pairwise interactions, we learn a boundary measure which defines the spatial discontinuity of object classes across an image. To model higher-order interactions, the proposed energy treats smooth surfaces as cliques and encourages all the pixels on a surface to take the same label. We show that the proposed higher-order energies can be decomposed into pairwise sub-modular energies and efficient inference can be made using the graph-cuts algorithm. We follow a systematic approach which uses structured learning to fine-tune the model parameters. We rigorously test our approach on SUN3D and both versions of the NYU-Depth database. Experimental results show that our work achieves superior performance to state-of-the-art scene labeling techniques. [ABSTRACT FROM AUTHOR]

Abstract: Industrial automation with speech control functions is generally installed with a speech recognition sensor which is used as an interface for users to articulate speech commands. However, recognition errors are likely to be produced when background noise surrounds the command spoken into the speech recognition microcontrollers. In this paper, a speech enhancement strategy is proposed to develop noise suppression filters in order to improve the accuracy of speech recognition microcontrollers. It uses a universal estimator, namely a neural network, to enhance the recognition accuracy of microcontrollers by integrating better signals processed by various noise suppression filters, where a global optimization algorithm, namely an intelligent particle swarm optimization, is used to optimize the inbuilt parameters of the neural network in order to maximize accuracy of speech recognition microcontrollers working within noisy environments. The proposed approach overcomes the limitations of the existing noise suppression filters intended to improve recognition accuracy. The performance of the proposed approach was evaluated by a speech recognition microcontroller, which is used in electronic products with speech control functions. Results show that the accuracy of the speech recognition microcontroller can be improved using the proposed approach, when working under low signal to noise ratio conditions in the industrial environments of automobile engines and factory machines. [Copyright &y& Elsevier]

A novel approach, using low Earth orbit (LEO) satellite microwave communication links for cloud liquid water measurements, is proposed in this paper. The feasibility of this approach is studied through simulations of the retrieval system including a LEO satellite communicating with a group of ground receivers equipped with signal-to-noise ratio (SNR) estimators, a synthetic cloud attenuation field and a tomographic retrieval algorithm. Rectangular and Gaussian basis functions are considered to define the targeted field. Simulation results suggest that the proposed least-squares based retrieval algorithm produces satisfactory outcomes for both types of basis functions. The root-mean-square error of the retrieved field is around 0.2 dB/km, with the range of the reference field as 0 to 2.35 dB/km. It is also confirmed that the partial retrieval of the cloud field is achievable when a limited number of receivers with restricted locations are available. The retrieval outcomes exhibit properties of high resolution and low error, indicating that the proposed approach has great potential for cloud observations. [ABSTRACT FROM AUTHOR]

This paper presents the development of a new model reduction method for discrete-time bilinear systems based on the balanced truncation framework. In many model reduction applications, it is advantageous to analyze the characteristics of the system with emphasis on particular frequency intervals of interest. In order to analyze the degree of controllability and observability of discrete-time bilinear systems with emphasis on particular frequency intervals of interest, new generalized frequency interval controllability and observability gramians are introduced in this paper. These gramians are the solution to a pair of new generalized Lyapunov equations. The conditions for solvability of these new generalized Lyapunov equations are derived and a numerical solution method for solving these generalized Lyapunov equations is presented. Numerical examples which illustrate the usage of the new generalized frequency interval controllability and observability gramians as part of the balanced truncation framework are provided to demonstrate the performance of the proposed method. [ABSTRACT FROM PUBLISHER]