• 말소리와 음성과학
• /
• 제10권1호
• /
• pp.39-48
• /
• 2018

A typical statistical parametric speech synthesis (text-to-speech, TTS) system consists of separate modules, such as a text analysis module, an acoustic modeling module, and a speech synthesis module. This causes two problems: 1) expert knowledge of each module is required, and 2) errors generated in each module accumulate passing through each module. An end-to-end TTS system could avoid such problems by synthesizing voice signals directly from an input string. In this study, we implemented an end-to-end Korean TTS system using Google's Tacotron, which is an end-to-end TTS system based on a sequence-to-sequence model with attention mechanism. We used 4392 utterances spoken by a Korean female speaker, an amount that corresponds to 37% of the dataset Google used for training Tacotron. Our system obtained mean opinion score (MOS) 2.98 and degradation mean opinion score (DMOS) 3.25. We will discuss the factors which affected training of the system. Experiments demonstrate that the post-processing network needs to be designed considering output language and input characters and that according to the amount of training data, the maximum value of n for n-grams modeled by the encoder should be small enough.

• 한국생산제조학회지
• /
• 제22권6호
• /
• pp.921-930
• /
• 2013

From an economic perspective, an enterprise's business activity depends on the efficient use of corporate resources for generating profits. However, on the enterprise side, it is difficult to measure and evaluate the effective use of each resource. This paper suggests an alternative for eliminating process inefficiencies in the consolidation of competitive power in auto parts manufacturing company A. Multitudinous process variables from company A's raw materials-to-shipment process are configured as input resources, and a Data Envelopment Analysis(DEA) is carried out to determine economical benefit of said resources' operation, as well as how products are manufactured. The DEA model offers a non-parametric approach to measuring relative efficiency using input and output factors. Furthermore, AHP is used for logically deciding the importance of each evaluation factor. In general, DEA models have been used for measuring efficiency of the service and public sectors. However, this study focused on measuring the efficiency of SMEs production lines.

• 한국환경과학회지
• /
• 제20권8호
• /
• pp.1011-1019
• /
• 2011

The objective of this research is to evaluate optimal conditions of permeability and selectivity on the polysulfone membrane for efficiency of separation of $CH_4$ by checking four factors which are temperature, pressure, gas compositions and gas flow rates. When higher pressure was applied at the input, lower efficiency of recovery of $CH_4$ and higher efficiency of separation of $CH_4$ were shown. It has the tendency to show lower efficiency of recovery of $CH_4$ and higher efficiency of separation of $CH_4$ at the output as higher temperature at input. The lower flow rates make higher efficiency of recovery of $CH_4$ and lower efficiency of separation of $CH_4$. Finally, over 90% efficiency for $CH_4$ separation and recovery conditions are temperature ($-5^{\circ}C$), pressure (8 bar), gas composition rate (6:4) ($CH_4:CO_2$) and gas flow rate ($5\ell$/min). These conditions make higher separation and recovery efficiency such as 90.1% and 92.1%, respectively.

• Asian-Australasian Journal of Animal Sciences
• /
• 제18권8호
• /
• pp.1194-1198
• /
• 2005

A computer vision system was designed and validated to recognize an individual Holstein cattle by processing images of their body patterns. This system involves image capture, image pre-processing, algorithm processing, and an artificial neural network recognition algorithm. Optimum management of individuals is one of the most important factors in keeping cattle healthy and productive. In this study, an image-processing system was used to recognize individual Holstein cattle by identifying the body-pattern images captured by a charge-coupled device (CCD). A recognition system was developed and applied to acquire images of 49 cattles. The pixel values of the body images were transformed into input data comprising binary signals for the neural network. Images of the 49 cattle were analyzed to learn input layer elements, and ten cattles were used to verify the output layer elements in the neural network by using an individual recognition program. The system proved to be reliable for the individual recognition of cattles in natural light.

• 소성∙가공
• /
• 제29권4호
• /
• pp.218-228
• /
• 2020

The prediction of final mass and optimized process conditions of injection molded products using Artificial Neural Network (ANN) were demonstrated. The ANN was modeled with 10 input parameters and one output parameter (mass). The input parameters, i.e.; melt temperature, mold temperature, injection speed, packing pressure, packing time, cooling time, back pressure, plastification speed, V/P switchover, and suck back were selected. To generate training data for the ANN model, 77 experiments based on the combination of orthogonal sampling and random sampling were performed. The collected training data were normalized to eliminate scale differences between factors to improve the prediction performance of the ANN model. Grid search and random search method were used to find the optimized hyper-parameter of the ANN model. After the training of ANN model, optimized process conditions that satisfied the target mass of 41.14 g were predicted. The predicted process conditions were verified through actual injection molding experiments. Through the verification, it was found that the average deviation in the optimized conditions was 0.15±0.07 g. This value confirms that our proposed procedure can successfully predict the optimized process conditions for the target mass of injection molded products.

• Design & Manufacturing
• /
• 제14권3호
• /
• pp.1-7
• /
• 2020

This paper predicts the mass and the length of injection-molded products through the Artificial Neural Network (ANN) method. The ANN was implemented with 5 input parameters and 2 output parameters(mass, length). The input parameters, such as injection time, melt temperature, mold temperature, packing pressure and packing time were selected. 44 experiments that are based on the mixed sampling method were performed to generate training data for the ANN model. The generated training data were normalized to eliminate scale differences between factors to improve the prediction performance of the ANN model. A random search method was used to find the optimized hyper-parameter of the ANN model. After the ANN completed the training, the ANN model predicted the mass and the length of the injection-molded product. According to the result, average error of the ANN for mass was 0.3 %. In the case of length, the average deviation of ANN was 0.043 mm.

• 가족자원경영과 정책
• /
• 제4권1호
• /
• pp.95-111
• /
• 2000

The family financial management is the important element which has a effect on the improvement of family economic well-being. Thus the study sets up human & material resources and requirement as an input factor, family financial management as an throughput factor, and economic well-being as an output factor, and then identifies if there are some relationships among them by using a systematic approach of family financial management. The questionaires were distributed to 600 employed wives by carrying out cluster sampling and disproportional stratified sampling. The 573 cases of them were used as the mean, ratio, Pearson’s Correration, ANOVA, factors analysis, and Stepwise Regression using the SAS/PC+. The major findings of the study are as follows: 1) A practical application of information as on input variable, appears the fator which has an important effect upon the throughput factor, that is, family financial management. The employed wives should manage a lot of resources in rapidly changing social-economic circustances. So it shows that not only they should take informations open to their daily life and then make practical use of them in managing their family finance, but also they must be capable of judging and accepting reasonable informations in order to gain and manage resources suitable to the characteristics of family finance and family members’need. 2) There was significant relationship between the employed wives’family financial management and family economic well-being, focused on the systematic theory of family financial management. This study provides for the information to develope the program for the employed wives’family financial planning and counseling.

• 한국기후변화학회지
• /
• 제6권2호
• /
• pp.121-131
• /
• 2015

As the temperature has changed by climate change, changes in its own characteristic values of the chemical substance or the movement and distribution of chemicals take place in accordance with the changes of hydrological and meteorological phenomena. Depending on the impact of climate change on the chemical behavior, it is necessary to understand and predict quantitative changes in the dynamics of the environment of pollutants due to climate change in order to predict in advance the occurrence of environmental disasters, and minimize the impact on the life and the environment after the incident. In this study, we have analysed and compared chemical fate models validated by previous studies in terms of model configuration, application size and input/output factors. The potential models applicable to municipal and industrial areas were selected on the basis of characteristic of each model, availability of input parameters and consideration for climate change, identified the problems, and then presented an approach to improve applicability.

• 농촌지도와개발
• /
• 제17권1호
• /
• pp.25-43
• /
• 2010

The objectives of this study are to estimate the number of employed persons related to agriculture in agribusiness, to determine any factors affecting changes in employed persons, based on the Input-Output Tables in 1995, 2000, and 2005. The number of employed persons related to Crops in the input industry of agriculture was 184,076 in 1995, 186,616 in 2000, and 193,394 in 2005. The sectors with higher number of employed persons were in; Agriculture, forestry and fishing related services, Wholesale and retail trade, Fertilizers and agricultural chemicals, and Forest products, etc. The number of employed persons related to Animals in the input industry of agriculture was 46,6166 in 1995, 28,886 in 2000, and 35,714 in 2005. The sectors with higher number of employed persons were in; Wholesale and retail trade, Prepared livestock feeds, Agriculture, forestry and fishing related services, and Forest products, etc. The number of employed persons related to Crops in post-farm gate industries was 184,076 in 1995, 186,616 in 2000, and 193,394 in 2005. The sectors with higher number of employed persons were in; Accommodation and food services, Other food products, Wholesale and retail trade, and Polished grains, flour and milled cereals, etc. The number of employed persons related to Animals in the post-farm gate industries was 184,076 in 1995, 186,616 in 2000, and 193,394 in 2005. The sectors with higher number of employed persons were in; Meat and dairy products, Wholesale and retail trade, Accommodation and food services, and Land transport, etc. In 2005, the number of employed persons related to agriculture in agribusiness per 1,000 employed persons in agriculture was 211, in the input industries in agriculture it was 53, and in the post-farm gate industries it was 158.gribusiness.

• Journal of Power Electronics
• /
• 제19권5호
• /
• pp.1108-1121
• /
• 2019

The non-inverting buck-boost converter (NIBB) is a step-up and step-down DC-DC converter suitable for wide-input-voltage-range applications. However, when the input voltage is close to the output voltage, the NIBB needs to operate in the buck-boost mode, causing a significant efficiency reduction since all four switches operates in the PWM mode. Considering both the current stress limitation and the efficiency optimization, a novel design methodology for the optimal phase-shift modulation of a NIBB in the buck-boost mode is proposed in this paper. Since the four switches in the NIBB form two bridges, the shifted phase between the two bridges can serve as an extra degree of freedom for performance optimization. With general phase-shift modulation, the analytic current expressions for every duty ratio, shifted phase and input voltage are derived. Then with the two key factors in the NIBB, the converter efficiency and the switch current stress, taken into account, an objective function with constraints is derived. By optimizing the derived objective function over the full input voltage range, an offline design methodology for the optimal modulation scheme is proposed for efficiency optimization on the premise of current stress limitation. Finally, the designed optimal modulation scheme is implemented on a DSPs and the design methodology is verified with experimental results on a 300V-1.5kW NIBB prototype.