Training and search methods for speech recognition.

Speech recognition involves three processes: extraction of acoustic indices from the speech signal, estimation of the probability that the observed index string was caused by a hypothesized utterance segment, and determination of the recognized utterance via a search among hypothesized alternatives. This paper is not concerned with the first process. Estimation of the probability of an index string involves a model of index production by any given utterance segment (e.g., a word). Hidden Markov models (HMMs) are used for this purpose [Makhoul, J. & Schwartz, R. (1995) Proc. Natl. Acad. Sci. USA 92, 9956-9963]. Their parameters are state transition probabilities and output probability distributions associated with the transitions. The Baum algorithm that obtains the values of these parameters from speech data via their successive reestimation will be described in this paper. The recognizer wishes to find the most probable utterance that could have caused the observed acoustic index string. That probability is the product of two factors: the probability that the utterance will produce the string and the probability that the speaker will wish to produce the utterance (the language model probability). Even if the vocabulary size is moderate, it is impossible to search for the utterance exhaustively. One practical algorithm is described [Viterbi, A. J. (1967) IEEE Trans. Inf. Theory IT-13, 260-267] that, given the index string, has a high likelihood of finding the most probable utterance.

Human Motion Analysis Based on Sequential Modeling of Radar Signal and Stereo Image Features

Falls are one of the greatest threats to elderly health in their daily living routines and activities. Therefore, it is very important to detect falls of an elderly in a timely and accurate manner, so that immediate response and proper care can be provided, by sending fall alarms to caregivers. Radar is an effective non-intrusive sensing modality which is well suited for this purpose, which can detect human motions in all types of environments, penetrate walls and fabrics, preserve privacy, and is insensitive to lighting conditions. Micro-Doppler features are utilized in radar signal corresponding to human body motions and gait to detect falls using a narrowband pulse-Doppler radar. Human motions cause time-varying Doppler signatures, which are analyzed using time-frequency representations and matching pursuit decomposition (MPD) for feature extraction and fall detection. The extracted features include MPD features and the principal components of the time-frequency signal representations. To analyze the sequential characteristics of typical falls, the extracted features are used for training and testing hidden Markov models (HMM) in different falling scenarios. Experimental results demonstrate that the proposed algorithm and method achieve fast and accurate fall detections. The risk of falls increases sharply when the elderly or patients try to exit beds. Thus, if a bed exit can be detected at an early stage of this motion, the related injuries can be prevented with a high probability. To detect bed exit for fall prevention, the trajectory of head movements is used for recognize such human motion. A head detector is trained using the histogram of oriented gradient (HOG) features of the head and shoulder areas from recorded bed exit images. A data association algorithm is applied on the head detection results to eliminate head detection false alarms. Then the three dimensional (3D) head trajectories are constructed by matching scale-invariant feature transform (SIFT) keypoints in the detected head areas from both the left and right stereo images. The extracted 3D head trajectories are used for training and testing an HMM based classifier for recognizing bed exit activities. The results of the classifier are presented and discussed in the thesis, which demonstrates the effectiveness of the proposed stereo vision based bed exit detection approach.

Modelling and testing volatility spillovers in oil and financial markets for USA, UK and China

The primary purpose of the paper is to analyze the conditional correlations, conditional covariances, and co-volatility spillovers between international crude oil and associated financial markets. The paper investigates co-volatility spillovers (namely, the delayed effect of a returns shock in one physical or financial asset on the subsequent volatility or co-volatility in another physical or financial asset) between the oil and financial markets. The oil industry has four major regions, namely North Sea, USA, Middle East, and South-East Asia. Associated with these regions are two major financial centers, namely UK and USA. For these reasons, the data to be used are the returns on alternative crude oil markets, returns on crude oil derivatives, specifically futures, and stock index returns in UK and USA. The paper will also analyze the Chinese financial markets, where the data are more recent. The empirical analysis will be based on the diagonal BEKK model, from which the conditional covariances will be used for testing co-volatility spillovers, and policy recommendations. Based on these results, dynamic hedging strategies will be suggested to analyze market fluctuations in crude oil prices and associated financial markets.

Exploring How AFFECT (A Family Focused Emotion Communication Training) and Self Psychology Are Mutually Enhancing

This paper implicitly advocates for a rapprochement between psychodynamic and behavioral approaches to psychotherapy, by exploring the similarities and differences between self psychology and A Family Focused Emotion Communication Training (AFFECT), a behavioral parent training model. Self psychology, a theory with broad applicability, has been applied to several modalities besides behavioral ones. Generally speaking, self psychology and AFFECT are both relational approaches to psychotherapy that emphasize the impact of parent responsiveness, more specifically empathic attunement, on a child's emotional development and emotion regulation. Differentiating aspects of each model are identified to enhance the other model. AFFECT has relevance for pushing self psychology theory more in the direction of operations, which has implications for enhancing the research potential of self psychology, as well as for the training of the self-psychologist. Conversely, self psychology has relevance for coaching the parent with low self-esteem and decreased self-efficacy in AFFECT, which has potential implications for AFFECT treatment outcomes.

Industry Professionals Who Attain Leadership in Energy and Environmental Design (LEED) Accreditation: An Examination of Motivations in Human Resource (HR) Training and Development

Despite the economy, the green building industry continues to grow and drive the demand for environmentally conscious, highly skilled professionals (USGBC 2009). LEED Accredited Professionals (APs) have the knowledge and skills to meet such demand; however, information is limited regarding LEED APs or their motivations and expectations toward prospective employers. The author surveyed a sample of LEED Accredited architects and found a combination of job and personal factors motivated them to attain accreditation. LEED APs value both a competitive salary and commitment to sustainability in prospective employers. To attract, retain, and utilize LEED APs, executives in this industry must reexamine corporate culture, their willingness to pay for credentialing, and the alignment of their reputation with the desires of potential applicants.

Manager Training and Development Strategy: A Methodology to Link Strategic Initiatives at Washington and Lee University

Washington and Lee University has developed and implemented a strategic plan and performance development initiative for 2007-2017. In concert with the strategic plan and performance development initiative, supportive manager core competencies have been identified. The deliverable for this capstone project is a documented methodology that supports the strategy for the design and implementation of a manager training and development program that ensures needed competencies are available. The author uses survey data to determine training needs and priorities and, through a review of literature, investigates effective strategies to arrive at a successful implementation methodology. The author presents findings and conclusions regarding design implementation methodology for manager training and development.

Preparation, characterisation and testing of CuO/Ce0.8Zr0.2O2 catalysts for NO oxidation to NO2 and mild temperature diesel soot combustion

CuO/ceria-zirconia catalysts have been prepared, deeply characterised (N2 adsorption–desorption isotherms at −196 °C, XRD, Raman spectroscopy, XPS, TEM and H2-TPR) and tested for NO oxidation to NO2 in TPR conditions, and for soot combustion at mild temperature (400 °C) in a NOx/O2 stream. The behaviour has been compared to that of a reference Pt/alumina commercial catalyst. The ceria-zirconia support was prepared by the co-precipitation method, and different amounts of copper (0.5, 1, 2, 4 and 6 wt%) were loaded by incipient wetness impregnation. The results revealed that copper is well-dispersed onto the ceria-zirconia support for the catalysts with low copper loading and CuO particles were only identified by XRD in samples with 4 and 6% of copper. A very low loading of copper increases significantly the activity for the NO oxidation to NO2 with regard to the ceria-zirconia support and an optimum was found for a 4% CuO/ceria-zirconia composition, showing a very high activity (54% at 348 °C). The soot combustion rate at 400 °C obtained with the 2% CuO/ceria-zirconia catalyst is slightly lower to that of 1% Pt/alumina in terms of mass of catalyst but higher in terms of price of catalyst.