Emotion speech recognition is a developing field in machine learning. The main purpose of this field is to produce a convenient system that is able to effortlessly communicate and interact with humans. The reliability of the current speech emotion recognition systems is far from being achieved. However, this is a challenging task due to the gap between acoustic features and human emotions, which rely strongly on the discriminative acoustic features extracted for a given recognition task. The speech signals were process with information which is divided into two main categories, linguistic and paralinguistic; emotions belong to the latter tree. The aim of this work is to develop a system that can understand paralinguistic information for paramount better human-machine interactions. A different extracted features like MFCC as well as feature classifications methods like HMM, GMM, LTSTM and ANN were used. In this paper, an improved architecture of CNN for speech emotion recognition were implemented. The main finding that the proposed CNN model achieved 93.96% accuracy rate in detecting emotions.
Deep learning is an emerging technology that is one of the most promising areas of artificial intelligence. Great strides have been made in recent years which resulted in increased efficiency with regards to many applications, including speech. Despite this, an environmentally Robust Speech Recognition system is still far from being achieved. In this article, an investigation of previous work has been conducted. The use of deep learning in speech recognition was analyzed and a proper deep learning architecture was identified. A method using convolutional neural network (CNN) is used with the aim of enhancing the performance of speech recognition systems (SRS). This study found that this CNN-based approach achieves a 94.32% validated accuracy
Accurate aiming of heliostats is critical to the successful performance of Concentrated Solar Power Tower Plants. The methodology employed at the Crescent Dunes Solar Plant to correct heliostat aiming utilizes images taken of heliostats aimed at a the center of a target. The centroids of those images are used as input to an estimation algorithm which calculates parameters used to correct the aiming. A significant problem is the presence of bad data in some of the input measurements. The Electric Power Industry has developed algorithms to detect and eliminate bad data in calculating the state variables of the power system. At Crescent Dunes we have adapted those algorithms to the problem of calculating the heliostat aiming correction parameters. This paper reports our experience to date with those algorithms.
In energy-constrain wireless sensor networks, maintaining k — coverage degree requested by an application while maximizing the network lifetime is a major challenge. Existing literature on k — coverage does not consider residual energy levels and actual battery discharge rate of ACTIVE nodes. OBSP (Optimized Backoff Sleep Protocol) considers the residual energy level information and battery discharge rate but ignores the k-coverage degree. This paper proposes k-CGP (k-Coverage Guarantee Protocol) based on battery discharge curve using polynomial regression for different coverage degrees (k). k-CGP determines the optimal wakeup rate of sleeping nodes by computing Optimal Sleep Time derived from battery discharge curve using polynomial regression and Received Signal Strength Indicator for distance estimation. The coverage redundancy is computed by using equi-distance test. Due to this, a sufficient number of ACTIVE nodes could be maintained while achieving lesser energy consumption in the network. Simulation results show that k-CGP achieves higher energy savings and sensing area coverage as compared to PEAS.
This paper presents a new design for aperture bowtie antenna with added rectangular and U shape metal strips to function at new different frequencies: 8.2, 9.25 and 10.5 GHz. The advantage of the proposed design is to use the original antenna shape and its dimensions to achieve the multiple frequencies. The proposed design consists of adding 2 U shape metal strips of similar sizes as well as 4 rectangular metal strips of different sizes for the original antenna. HFSS software is used in the simulation of the antenna in order to obtain reflection coefficients, radiation patterns, impedances and surface current distributions.
