This article addresses the robust estimation of the output layer linear parameters in a radial basis function network (RBFN). A prominent method used to estimate the output layer parameters in an RBFN with the predetermined hidden layer parameters is the least-squares estimation, which is the maximum-likelihood (ML) solution in the specific case of the Gaussian noise. We highlight the connection between the ML estimation and minimizing the Kullback-Leibler (KL) divergence between the actual noise distribution and the assumed Gaussian noise. Based on this connection, a method is proposed using a variant of a generalized KL divergence, which is known to be more robust to outliers in the pattern recognition and machine-learning problems. The proposed approach produces a surrogate-likelihood function, which is robust in the sense that it is adaptive to a broader class of noise distributions. Several signal processing experiments are conducted using artificially generated and real-world data. It is shown that in all cases, the proposed adaptive learning algorithm outperforms the standard approaches in terms of mean-squared error (MSE). Using the relative increase in the MSE for different noise conditions, we compare the robustness of our proposed algorithm with the existing methods for robust RBFN training and show that our method results in overall improvement in terms of absolute MSE values and consistency.

Principal component analysis (PCA) is widely used for feature extraction and dimension reduction in pattern recognition and data analysis. Despite its popularity, the reduced dimension obtained from the PCA is difficult to interpret due to the dense structure of principal loading vectors. To address this issue, several methods have been proposed for sparse PCA, all of which estimate loading vectors with few non-zero elements. However, when more than one principal component is estimated, the associated loading vectors do not possess the same sparsity pattern. Therefore, it becomes difficult to determine a small subset of variables from the original feature space that have the highest contribution in the principal components. To address this issue, an adaptive block sparse PCA method is proposed. The proposed method is guaranteed to obtain the same sparsity pattern across all principal components. Experiments show that applying the proposed sparse PCA method can help improve the performance of feature selection for image processing applications. We further demonstrate that our proposed sparse PCA method can be used to improve the performance of blind source separation for functional magnetic resonance imaging data.

Data-driven methods, such as principal component analysis and independentcomponent analysis, have been successfully applied to functionalmagnetic resonance imaging (fMRI) data in particular and neuro-imaging data in general. A central issue of thesemethods is the importance of correctly selecting the number of components to be used in the factor model. This issue is often addressed using a model selection criterion, where the goodness-of-fit term is obtained from the log-likelihood function. In this paper, an alternative criterion is proposed for selecting the number of components. Unlike existingmodel selection criteria that use the log-likelihood function, the proposed goodness-of-fit termuses the sum of squares of the smallest eigenvalues of the sample covariance matrix. The proposed criterion is obtained from the asymptotic distribution of the goodness-of-fit term, for which consistency is established. This criterion has a straight-forward implementation and is shown to outperform conventional model selection criteria used in fMRI data analysis. Experiments are conducted using simulated and real fMRI data, in which improved performance is obtained by the proposed criterion, both in terms of accuracy and consistency under data variabilities.

We address the problem of two-dimensional (2-D) whitening of face image matrices. The proposed method whitens the distribution of rows and columns of an image matrix. The main contribution of this letter is to investigate some aspects of recently published studies on this topic. We point out that existing methods on 2-D whitening for face recognition overlook the rows as potential random vectors, implying that the image matrix is only a collection of independent identically distributed column vectors. This study shows that this one-sided whitening approach relies on an incorrect assumption for image data. We show that one-sided whitening replaces the covariance matrix along one dimension with the identity matrix. Our proposed, truely 2-D, whitening transform does not enforce any such restriction on the row- or column-covariance matrices. A second contribution of our letter is to illustrate a method to confirm that the rows and columns of front-pose face images are in fact approximately normally distributed. This validation has not been addressed previously, although it plays a crucial role in deriving a linear whitening transform.

This document briefly describes the systems submitted by the Center for Robust Speech Systems (CRSS) from The University of Texas at Dallas (UTD) to the 2016 National Institute of Standards and Technology (NIST) Speaker Recognition Evaluation (SRE). We developed several UBM and DNN i-Vector based speaker recognition systems with different data sets and feature representations. Given that the emphasis of the NIST SRE 2016 is on language mismatch between training and enrollment/test data, so-called domain mismatch, in our system development we focused on: (1) using unlabeled in-domain data for centralizing data to alleviate the domain mismatch problem, (2) finding the best data set for training LDA/PLDA, (3) using newly proposed dimension reduction technique incorporating unlabeled in-domain data before PLDA training, (4) unsupervised speaker clustering of unlabeled data and using them alone or with previous SREs for PLDA training, (5) score calibration using only unlabeled data and combination of unlabeled and development (Dev) data as separate experiments.

Overlapped speech is referred to a monophonic audio signal in which at least two speakers are present at the same time. In this study, the focus is on distinguishing overlapped from single-speaker speech, i.e., overlapped speech detection. We develop an overlap detection algorithm using an enhanced time-frequency representation, called Pyknogram, estimated directly from the input audio signal. Pyknograms use the Teager-Kaiser energy operator to detect resonant time-frequency units and thereby suppress nonharmonic structures. We show how the resulting Pyknograms provide high separability in terms of detecting the presence of interfering speech. Our proposed unsupervised Pyknogram-based detection results in over 30% relative improvement in overlap detection error rates across different signal-to-interference ratios (SIR) compared to baseline systems. In addition, a case study is presented where we evaluate speaker verification performance under different overlap conditions using the GRID database and observe that speaker verification equal error rates (EER) vary from 2% to 30%, depending on the average SIR values introduced to train and test sets. In order to estimate the reliability of speaker verification scores across different trials, overlap detection results are interpreted as low-level information and stacked alongside verification outputs. The resulting high-dimensional space is passed through a support vector machine classifier to find the separating hyperplane between target and imposter scores. Combining overlap detection scores with speaker verification on average yields 20% relative decrease in EER. We also provide an upper bound for this approach using existing overlap labels, which yields 23% relative improvement.

The ability to estimate the number of words spoken by an individual over a certain period of time is valuable in second language acquisition, healthcare, and assessing language development. However, establishing a robust automatic framework to achieve high accuracy is non-trivial in realistic/naturalistic scenarios due to various factors such as different styles of conversation or types of noise that appear in audio recordings, especially in multi-party conversations. In this study, we propose a noise robust overlapped speech detection algorithm to estimate the likelihood of overlapping speech in a given audio file in the presence of environment noise. This information is embedded into a word-count estimator, which uses a linear minimum mean square estimator (LMMSE) to predict the number of words from the syllable rate. Syllables are detected using a modified version of the mrate algorithm. The proposed word-count estimator is tested on long duration files from the Prof-Life-Log corpus. Data is recorded using a LENA recording device, worn by a primary speaker in various environments and under different noise conditions. The overlap detection system significantly outperforms baseline performance in noisy conditions. Furthermore, applying overlap detection results to word-count estimation achieves 35% relative improvement over our previous efforts, which included speech enhancement using spectral subtraction and silence removal.

Novice young drivers are more frequently involved in traffic accidents, and studies have shown that effective supervised driver training is the key in reducing young drivers' risks. Using our previously developed Mobile-UTDrive in-vehicle data acquisition platform, two 16-age novice drivers participated in naturalistic drive training data collection. This paper focuses on analysis of novice driver training signals from an audio processing perspective. Specifically, analysis of supervised driver instruction audio and resulting CAN-Bus maneuver operation is performed. Following a procedure which consists of noise suppression, speech recognition and keyword spotting, five tutorial keywords - Brake, Gas, Left, Right and Stop - are spotted at an overall accuracy rate of 40% versus all spontaneous continuous speech. The time stamps of these keywords are then used as indications of driving maneuvers. As examples of driving performance evaluation, the case of making Left-Turn maneuvers for the two novice drivers are assessed and compared, and the increase of driving skills over experiences are analyzed.

State-of-the-art speaker verification systems model speaker identity by mapping i-Vectors onto a probabilistic linear discriminant analysis (PLDA) space. Compared to other modeling approaches (such as cosine distance scoring), PLDA provides a more efficient mechanism to separate speaker information from other sources of undesired variabilities and offers superior speaker verification performance. Unfortunately, this efficiency is obtained at the cost of a required large corpus of labeled development data, which is too expensive/unrealistic in many cases. This study investigates a potential solution to resolve this challenge by effectively utilizing unlabeled development data with universal imposter clustering. The proposed method offers +21.9% and +34.6% relative gains versus the baseline system on two public available corpora, respectively. This significant improvement proves the effectiveness of the proposed method.

This paper describes the systems developed by the Center for Robust Speech Systems (CRSS), for the 2012 National Institute of Standards and Technology (NIST) Speaker Recognition Evaluation (SRE). Given that the emphasis of SRE'12 is on noisy and short duration test conditions, our system development focused on: (i) novel robust acoustic features, (ii) new feature normalization schemes, (iii) various back-end strategies utilizing multi-session and multi-condition training, and (iv) quality measure based system fusion. Noisy and short duration training/test conditions are artificially generated and effectively utilized. Active speech duration and signal-to-noise-ratio (SNR) estimates are successfully employed as quality measures for system calibration and fusion. Overall system performance was very successful for the given test conditions.

In this study we propose a system for overlapped-speech detection. Spectral harmonicity and envelope features are extracted to represent overlapped and single-speaker speech using Gaussian mixture models (GMM). The system is shown to effectively discriminate the single and overlapped speech classes. We further increase the discrimination by proposing a phoneme selection scheme to generate more reliable artificial overlapped data for model training. Evaluations on artificially generated co-channel data show that the novelty in feature selection and phoneme omission results in a relative improvement of 10% in the detection accuracy compared to baseline. As an example application, we evaluate the effectiveness of overlapped-speech detection for vehicular environments and its potential in assessing driver alertness. Results indicate a good correlation between driver performance and the amount and location of overlapped-speech segments.

I4U is a joint entry of nine research Institutes and Universities across 4 continents to NIST SRE 2012. It started with a brief discussion during the Odyssey 2012 workshop in Singapore. An online discussion group was soon set up, providing a discussion platform for different issues surrounding NIST SRE’12. Noisy test segments, uneven multi-session training, variable enrollment duration, and the issue of open-set identification were actively discussed leading to various solutions integrated to the I4U submission. The joint submission and several of its 17 sub-systems were among top-performing systems. We summarize the lessons learnt from this large-scale effort.

In-vehicle conversations are typical when there is more than one person in the car. Although many conversations are beneficial in keeping the driver alert and active, there are also instances where a competitive conversation may adversely influence driving performance. Identifying such scenarios can improve vehicle safety systems by fusing the knowledge obtained from conversational speech analysis and vehicle dynamic signals. In this study we incorporate the use of smart portable devices to create a unified platform for recording in-vehicular conversations as well as the vehicle dynamic signals required to evaluate the driving performance. Results show that turn taking rate and overlapping speech segments under certain conditions correlate with deviations from normal driving patterns. The conversational speech analysis can thus be utilized as a component in driver assistance systems such that the impact of in-vehicle speech activity on driving performance is controlled or minimized.