• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1998.06a
• /
• pp.95-98
• /
• 1998

본 논문에서는 코퍼스 기반 음성합성시 각 음성 세그머트의 에너지 정규화가 합성된 음성의 음질에 미치는 영향에 대하여 연구한다. 음성합성에 사용되는 음성 세그먼트를 실제 자연 음성 데이터로부터 추출된 것으로 다양한 발음세기를 가진다. 따라서 이들을 조합하여 만든 합성음성의 음질은 일반적으로 음량이 고르지 못하고 듣기에 부자연스럽다. 이러한 문제를 해결하기 위해 음성합성시 음성 세그먼트의 에너지를 정규화하는 방법을 제안하고 정규화방법으로 최대진폭 정규화방식을 사용하였다. 녹음환경이 비교적 일정한 코퍼스와 그렇지 않은 환경에서 녹음된 코퍼스를 사용하여 정규화 없이 합성한 음성의 음질과 정규화를 거쳐서 합성한 음성의 음질을 비교한다. 실험결과 음성 세그먼트의 에너지를 정규화한 경우 합성음성의 음질이 개선되었다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07b
• /
• pp.157-159
• /
• 2005

본 논문에서는 단일 색인을 사용하는 정규화 변환 지원 서브시퀀스 매칭 방법을 제안한다. 기존의 정규화 변환 지원 서브시퀀스 매칭 방법은 질의 시퀀스 길이가 커질수록 성능이 저하되고, 이를 해결하기 위하여 여러개의 색인을 사용하는 방법을 취하였다. 본 논문에서는 하나의 색인을 사용하면서도 다양한 길이의 정규화 변환 지원 서브시퀀스 매칭을 수행하는 효율적인 방법을 제시한다. 이를 위하여, 본 논문에서는 정규화 변환의 정의를 확장하여 일반화 정규화 변환 개념을 제시한다. 또한, 이러한 일반화 정규화 변환 개념을 기존 서브시퀀스 매칭 방법들에 적용하는 방안에 대한 이론적 근거를 각각의 정리로서 제시하고 증명하였다. 그리고, 이들 방안을 구현하기 위한 색인 구성 알고리즘 및 서브시퀀스 매칭 알고리즘을 각각 제시하였다. 본 논문에서 제안한 정규화 변환 지원 서브시퀀스 매칭은 다른 변환을 지원하는 서브시퀀스 매칭으로 일반화 될 수 있는 우수한 연구결과라 사료된다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.17 no.1
• /
• pp.41-48
• /
• 2022

This paper presents the unified Bayesian Tikhonov regularization method as a solution to total variation regularization. The integrated method presents a formula for obtaining the regularization parameter by transforming the total variation term into a weighted Tikhonov regularization term. It repeats until the reconstructed image converges to obtain a regularization parameter and a new weighting factor based on it. The experimental results show the effectiveness of the proposed method for the image restoration problem.

• Journal of Labour Economics
• /
• v.24 no.1
• /
• pp.207-230
• /
• 2001

This paper uses the Korean Labor Panel data to investigate changes in the employment types of male workers following their job changes with the classification of workers into three categories: regular wage workers, non-regular wage workers, and self-employed workers. It also estimates a competing-risks hazard model to analyze the determinants of employment types of workers. The results show that the type of employment of a worker at an immediate previous job has a critical importance in determining his employment type at a new job and that the types of employment at jobs other than the immediate previous job also play some role in determining the type of employment at a new job, although their impact declines as the number of intervening jobs increases. A job loser, who worked as a non-regular worker at his immediate previous job, for example, is considerably less likely to find a regular job, but more likely to get reemployed at another non-regular job than one who worked as a regular worker at his immediate previous job. Similarly, a worker who quit self-employment is much less likely to find a regular job but more likely to restart his own business than one who worked as a regular worker at his immediate previous job. These findings suggest that it is not easy at all for a worker who worked as either a non-regular worker or self-employed worker to become a regular worker, although it might be premature to assert that non-regular jobs or self-employed jobs are dead-end jobs. Another interesting finding of this analysis is that a high unemployment rate lowers a probability of reemployment at either regular jobs or self-employed jobs, but raises a non-regular job reemployment probability, which strongly implies that as labor market conditions become adverse to workers the proportion of non-regular employment can rise rapidly.

• The Journal of the Korea Contents Association
• /
• v.9 no.5
• /
• pp.314-321
• /
• 2009

The main purpose of this paper is to examine the relationship between employment relations characteristic and job involvement according to employment type in hotel corporations. Using data from 238 employees from hotel corporations, this research shows the results in the below. First, there is a significant difference in the employment relations characteristic(employment security and peer relationships) and job involvement. Second, the employment type does not moderates the relationships between employment relations characteristic and job involvement. Third, employment security and peer relationships are powerful factors to predict job involvement in both form of worker.

• Journal of Labour Economics
• /
• v.36 no.1
• /
• pp.63-91
• /
• 2013

This paper analysed the training effects on wage and employment security by the non-standard worker types by using Economically Active Population Additional Survey data. Through propensity score matching method, we found that the regular workers' training effect on wage was 4.2%, that was very higher than 2.7% of fixed-term workers. Logistic regression analysis showed that the training participation itself did not affect on the regular workers' regular position continuance, but training duration affected. Just in fixed-term workers among non-standard workers, both of training participation and training duration affected the transition from non-standard position to regular position. The result that both training effects on wage and on employment were positive just in the fixed-term workers might be interpreted owing to the clear employment contract relation between employer and employees.

• 한국사회정책
• /
• v.25 no.4
• /
• pp.125-161
• /
• 2018

This article aims to investigate the impact of Korean employment protection legislation that has implemented since 2007 on eliminating discrimination on non-regular worker's wage and social security. It is used the panel Tobit model reflecting the variation of implementation time according to the size of establishments. Although the employment protection laws for non-regular workers have implemented, the wage gap and discrimination in social security for non-regular workers have continued. Of course, the discrepancies on wage and social security were founded not only between regular and non-regular workers but also within non-regular workers. For reducing the discriminations, this study proposes to restrict the reason for justifying discrimination, and the introduction of a new approach to accessing the discrimination and complimentary credit system. Besides, this study suggests to actively review the strengthening of regulations on the use of non-regular workers.

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.42 no.5
• /
• pp.808-817
• /
• 2013

This study was conducted to compare job satisfaction and factors related to job satisfaction between permanently and temporarily employed dietitians in school foodservices in the Gyeongsangnam-do area. A total of 204 questionnaires were used in the final analysis. The average age, length of employment, and monthly wage of temporarily employed dietitians was significantly lower than those of permanently employed dietitians. However, there was no significant difference of overall job satisfaction between permanently and temporarily employed dietitians, although the average pay, welfare benefits, and promotion factors for permanently employed dietitians was significantly higher. Work and pay factors had significant effects on the overall job satisfaction of permanently employed dietitians, while work factors and work atmosphere had significant effects on the overall job satisfaction of temporarily employed dietitians. In conclusion, there was a significant difference in overall factors related to job satisfaction between permanently employed dietitian and temporarily employed dietitians. The pay, welfare benefits, and promotion condition of temporarily employed dietitians should be improved to ensure the efficient management of the school foodservice workforce in the future.

• Korean Journal of Labor Studies
• /
• v.19 no.1
• /
• pp.213-260
• /
• 2013

This study pays attention to the restricted use and the possibility of standard position of contingent workers in Brazil. The labor market of Brazil has been developed by formal and informal labor sector, and informal sector includes various precarious workers as well as contingent workers. According to Brazilian Statistics Department, not contingent workers but informal labor focused in this paper have been slowly decreased since year 2000. In this context, this study investigated on the reasons of decreasing contingent employment in Brazil. The results demonstrate that decreased informal employment and instead increased standard workers could not be interpreted by recent the Braizil's economic boom. Along with literature review, the author conducted the case study regarding employment of contingent workers at six large foreign companies in Sao Paulo. The results of this show that the use of contingent employment was prohibited in regular daily works by the labor law and thus firms employed contingent workers in only temporary positions. Further, firms often promise standard positions for contingent workers when temporary employment contract was terminated since there is little or no exist of the differences of wage between standard and contingent worker in terms of 'same work same wage' and 'minimum wage'. In here, labor unions play a key role in employment change from contingent position to standard job. Consequently, decreasing of contingent workers and stepping stone to regular jobs seems to be triggered by both legal regulation on contingent employment and strong unions. This institutional perspective may extend the theoretical view on the use of contingent workers, and the author discuss that Brazil's case could provide practical implications to Korean labor policy.

• Journal of Labour Economics
• /
• v.34 no.3
• /
• pp.119-139
• /
• 2011

Using a establishment-worker matched data, this paper estimates wage differentials between standard and non-standard workers. Unlike previous studies, we estimate a fixed-effect model for the tree-way error-components that control for both unobserved individual heterogeneities and unobserved firm heterogeneities. The estimation results show that standard workers earn 6.5~8.4% mire than non-standard workers. This wage premium is 30~40% of the wage differential estimated from the OLS model. The results implies that a large proportion of the wage differentials between standard and non standard workers can be explained by unobserved firm and individual characteristics.