• MALSORI
• /
• no.45
• /
• pp.35-45
• /
• 2003

Laryngeal diseases cause significant changes in the quality of speech production. Automatic detection of laryngeal diseases by voice is attractive because of its nonintrusive nature. In this paper, we apply speech recognition techniques to detection of laryngeal cancer, and investigate which feature parameters and classification methods are appropriate for this purpose. Linear Predictive Cepstral Coefficients (LPCC) and Mel-Frequency Cepstral Coefficients (MFCC) are examined as feature parameters, and parameters reflecting the periodicity of speech and its perturbation are also considered. As for classifier, multilayer perceptron neural networks and Gaussian Mixture Models (GMM) are employed. According to our experiments, higher order LPCC with the periodic information parameters yields the best performance.

• Journal of Multimedia Information System
• /
• v.3 no.2
• /
• pp.1-8
• /
• 2016

In this paper, we propose a method to automatically diagnose various diseases. The input data consists of electrocardiograph (ECG) recordings. We extract R-to-R interval (RRI) signals from ECG recordings, which are preprocessed to remove trends and ectopic beats, and to keep the signal stationary. After that, we perform some prospective analysis to extract time-domain parameters, frequency-domain parameters, and nonlinear parameters of the signal. Those parameters are unique for each disease and can be used as the statistical symptoms for each disease. Then, we perform feature selection to improve the performance of the diagnosis classifier. We utilize the selected features to diagnose various diseases using machine learning. We subsequently measure the performance of the machine learning classifier to make sure that it will not misdiagnose the diseases. The first two steps, which are R-to-R extraction and preprocessing, have been successfully implemented with satisfactory results.

• International Journal of Fluid Machinery and Systems
• /
• v.2 no.2
• /
• pp.172-178
• /
• 2009

This paper reports on an investigation (using RSM with commercial CFD software) of the performance characteristics of the impeller in a centrifugal pump. Geometric parameters of vane plane development were defined with the meridional shape and frontal view of the impeller. The parameters are focused on the blade-angle distributions through the impeller in a fixed meridional geometry. For screening, a $2^k$ factorial design has been used to identify the important design parameters. The objective functions are defined as the total head rise and the total efficiency at the design flow-rate. From the $2^k$ factorial design results, it is found that the incidence angles and the exit blade angle are the most important parameters influencing the performance of the pump.

• The Transactions of the Korean Institute of Electrical Engineers P
• /
• v.61 no.4
• /
• pp.199-205
• /
• 2012

This paper propose a improved parameter estimations of five-phase squirrel-cage induction motor(IM) for speed control system on field oriented control(FOC). In order to high performance control of ac the motors using a FOC and DTC(direct torque control) algorithm, there are required precise motor parameters for slip calculation, flux observer, controller gain, rotor position and speed estimation, and so on. We are suggest a estimation method of the motor parameters that developing five-phase squirrel-cage IM have a stator of concentrated winding for experimental. There are results of stator winding test, no-load test, locked rotor test, and obtained equivalent circuits using manufactured experimental apparatus. For presenting the superior performance of the speed control system in adapted the parameters, experimental results are presented using a 32-bit fixed point TMS320F2812 DSP with 1.5[KW] IM.

• Journal of Ship and Ocean Technology
• /
• v.7 no.2
• /
• pp.20-28
• /
• 2003

A table of annual sea state occurrences is prepared based on the significant wave heights and modal periods, two parameters representing a Bretschneider wave spectrum. Measurements of them were done by the Korean Ocean Research and Development Institute (KORDI) at 67points around the Korean peninsula for 22 years. Also the average values of parameters over 4 regions and 4 seasons were derived. Comparative seakeeping calculations using wave parameters prevalent around Korea and those of the North Pacific at sea state numbers 3, 4, and 5 respectively, show some differences in the assessed seakeeping performance.

• Journal of the Korean Ceramic Society
• /
• v.56 no.3
• /
• pp.325-330
• /
• 2019

Band convergence is known to be effective in improving thermoelectric performance by increasing the Seebeck coefficient without significantly reducing electrical conductivity. Decoupling of the Seebeck coefficient and electrical conductivity in converged bands is the key requirement. Yet, the degree of decoupling depends on the band parameters of the converging bands. Herein, we report theoretical transport properties of two valence bands as their energy difference changes from 0.25 eV to 0 eV. In order to demonstrate the effect of band parameters in transport, we first conducted calculations for the case where the two bands have the same parameters. Then, we conducted the same calculation by doubling the density-of-states effective mass of one valence band. Given that there are two bands, each band's effective mass was doubled one at a time while the other band's effective mass remained constant. We found that the decoupling was strongest when the bands participating in convergence had the same band parameters.

• ETRI Journal
• /
• v.41 no.1
• /
• pp.109-116
• /
• 2019

This paper explores new techniques that are based on a hidden-layer linear transformation for fast speaker adaptation used in deep neural networks (DNNs). Conventional methods using affine transformations are ineffective because they require a relatively large number of parameters to perform. Meanwhile, methods that employ singular-value decomposition (SVD) are utilized because they are effective at reducing adaptive parameters. However, a matrix decomposition is computationally expensive when using online services. We propose the use of an extended diagonal linear transformation method to minimize adaptation parameters without SVD to increase the performance level for tasks that require smaller degrees of adaptation. In Korean large vocabulary continuous speech recognition (LVCSR) tasks, the proposed method shows significant improvements with error-reduction rates of 8.4% and 17.1% in five and 50 conversational sentence adaptations, respectively. Compared with the adaptation methods using SVD, there is an increased recognition performance with fewer parameters.

• Asian-Australasian Journal of Animal Sciences
• /
• v.19 no.10
• /
• pp.1478-1483
• /
• 2006

This experiment was carried out to determine the effects of supplementation of L-carnitine and humic substances alone or in combination in laying hen diets on performance, egg traits and blood parameters. A total of 180 IGH type brown laying hens aged 22 weeks were employed in a completely randomized block design with one control group and three treatment groups. Each group was divided into five replicates as subgroups, each comprising 9 hens. The diets of the first, second and third treatment groups were supplemented with 0.1 g/kg L-carnitine, 1.5 g/kg humic substances (Farmagulator$^{(R)}$ Dry Plus) and 0.1 g/kg L-carnitine+1.5 g/kg humic substances, respectively. The experimental period lasted 18 weeks. Feeding supplemental carnitine, humic substances or carnitine+humic substances resulted in increases in body weight gain (p<0.05). Dietary treatments did not significantly affect daily feed intake, daily metabolizable energy intake, egg production, egg weight, feed efficiency, mortality, egg shape index, egg breaking strength, egg shell thickness, egg albumen index, egg yolk index, egg Haugh unit and the percentages of egg shell, albumen and yolk. Supplementation of humic substances reduced egg yolk cholesterol as mg per g yolk and mg per yolk (p<0.05). Blood serum parameters were not affected by the supplementation of carnitine, humic substances or carnitine+humic substances. The results in this study demonstrated that humic substances supplementation reduced egg cholesterol without adverse effects on performance, egg traits and blood parameters of laying hens. It was concluded that the usage of L-carnitine alone or in combination with humic substances in diets had no beneficial effects in laying hens.

• Journal of Korean Society of Water and Wastewater
• /
• v.31 no.1
• /
• pp.1-6
• /
• 2017

Surge pressure is created by rapid change of flow rate due to operation of hydraulic component or accident of pipeline. Proper control of surge pressure in distribution system is important because it can damage pipeline and may have the potential to degrade water quality by pipe leakage due to surge pressure. Surge relief valve(SRV) is one of the most widely used devices and it is important to determine proper parameters for SRV's installation and operation. In this research, determining optimum parameters affecting performance of the SRV were investigated. We proposed the methodology for finding combination of parameters for best performance of the SRV. Therefore, the objective function for evaluate fitness of candidate parameters and surge pressure simulation software was developed to validate proposed parameters for SRV. The developed software was integrated into genetic algorithm(GA) to find best combination of parameters.

• Korean Journal of Computational Design and Engineering
• /
• v.13 no.6
• /
• pp.440-449
• /
• 2008

This paper deals with a method calculating various hull form parameters which are required in numerical analysis for ship performance such as motion, maneuverability, resistance and propulsion, etc. After the hull form is designed, before the model tests the ship's performances are evaluated by various analysis tools in which the hull form parameters are used with many kinds of forms aside from offset data. Here, The hull form parameters characterize the properties of hull form and contain positional, differential and integral information implicitly. Generally, the commercial CAD-system has not functions enough for supporting these form parameters and therefore each shipyard uses its own in-house analysis program as well as commercial analysis software. To overcome these limitations, modules for supporting these analysis programs have developed. The modules contain cubic composite spline cure using local curve fairing, intersect algorithm, Gaussian integral, and other geometric techniques needed in calculating hull form parameters. Using our analysis-supporting modules, a complex hull form can be remodeled exactly to the hull form designed by CAD-system and any hull form parameter required in various performance analyses can be calculated.