• The Journal of the Acoustical Society of Korea
• /
• v.17 no.6
• /
• pp.37-42
• /
• 1998

본 논문에서는 SCHMM에 적용된 관찰 확률 최대화에 의한 화자 적응 알고리즘을 제안한다. 제안된 알고리즘은 SCHMM의 관찰 확률 밀도들이 새로운 화자의 음성 특징을 잘 표현하지 못하는 경우 인식 성능이 저하되는 것을 막기 위하여, 적응 데이터의 각 특징 벡터들이 최대의 관찰 확률을 가질 수 있도록 관찰 확률 밀도를 결정하는 평균 벡터 μ와 분산 행렬 Σ를 기울기 탐색(gradient search) 알고리즘에 의해 반복적으로 적응시켜 주는 방법이다. SCHMM의 상태 천이 확률 A와 혼합 밀도 계수 C는 관찰 확률 밀도 적응 과정 을 거친 후, 적응 데이터로부터 구한 확률과 기존 확률의 가중 평균을 취하는 과정을 반복 하여 적응시켜 주었다. 제안된 화자 적응 알고리즘을 사용하여 단독음 인식 실험을 수행한 결과, 화자 적응을 수행하지 않았을 때와 비교하여 화자 독립 시스템에서는 평균 9.8%, 남 성 화자 종속 시스템에서는 평균 46.0%, 여성 화자 종속 시스템에서는 평균 52.7%의 인식 률 향상을 보였다.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.4
• /
• pp.397-409
• /
• 2022

A model is developed for the calculation of sea surface generated ambient noise in the range dependent ocean environment. The sources are located in the horizontal plane all around and their depths are at the near-surface. The receiver array is located in the range dependent ocean waveguide. One-way coupled mode method is used to model the acoustic propagation between the sources and receiver in the range dependent waveguide, and the cross spectral density matrix of noise is derived. In simulation, noise intensity, beamforming result and coherence function are calculated from the cross spectral density matrix. These results are compared with those in the range independent environment. The modeling result shows the effect of the vertical directionality and asymmetry characteristics of the horizontal plane.

• Communications for Statistical Applications and Methods
• /
• v.16 no.4
• /
• pp.587-594
• /
• 2009

In this paper the change point problem of the error terms in linear regression models is considered. Since fixed or stochastic independent variables and weakly dependent errors are assumed, usual multiple regression models and time series models including ARMA are covered. We use the estimates of probability density function based on residuals in order to test the distribution change of the unobserved errors. Under some mild conditions, the test using the residuals is proved to have the same limiting distribution as the test based on true errors.

• Korean Journal of Fisheries and Aquatic Sciences
• /
• v.34 no.5
• /
• pp.423-429
• /
• 2001

Prediction of the variation in annual biomass was conducted for the white croaker. Argyrosomus argentatus in Korean waters using leslie Matrix, based upon fishery data for the past 21 years and biological data, We used density-independent and density-dependent Leslie Matrix models. Similar parameters were estimated from two models except that the density-dependent model was influenced by the density effect variable, q(i,t), The eigenvalue of the white croaker population for the $1984\~1995$ period was estimated to be 0.8, indicating a declining pattern of the population. The survival rate of 0-th year class was calculated to be 0.00005. Based on the schedule of the age-specific survival rate and fecundity, the future biomass and catch was predicted for various levels of fishing mortalities (F), If F was set at 0.252/yr ($F_{35x}$) or 0.368/yr ($F_{0.1}$), the biomass and catch increased, and if F was set at 0.922 ($F_{current}$), the biomass and catch decreased, The fishing mortality at equilibrium was estimated to be 0.7/yr. Finally, the management strategy of the white croaker was discussed.

• Proceedings of the Korea Water Resources Association Conference
• /
• 1999.02b
• /
• pp.601-652
• /
• 1999

• Applied Chemistry for Engineering
• /
• v.7 no.6
• /
• pp.1147-1155
• /
• 1996

In the spray drying with rotating wheel atomizer of cheese powder, the relationships among variables were analyzed with Response Surface Methodology in which several independent variables such as total solid content, wheel rotation speed, and outlet temperature influenced dependent variables such as particle diameter, moisture content, bulk density, and viscosity of suspended liquid. Significance and correlation were tested according to central composite design. As a results of analyzing the correlations between independent and dependent variables, particle diameter and moisture content of cheese powder were decreased with increasing wheel rotation speed, and bulk density was decreased with increasing outlet temperature. Viscosity of suspended liquid were increased with increasing wheel rotation speed and total moisture content. In correlation among dependent variables, moisture content was proportional to bulk density, and particle diameter was inversly proportional to moisture content and bulk density.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2006.11a
• /
• pp.175-178
• /
• 2006

본 논문에서는 주요성분분석과 상호정보 추정을 조합한 입력변수선택 기법을 제안하였다. 여기서 주요성분분석은 2차원 통계성을 이용하여 입력변수 간의 독립성을 찾기 위함이고, 상호정보의 추정은 적응적 분할을 이용하여 입력변수의 확률밀도함수를 계산함으로써 변수상호간의 종속성을 좀더 정확하게 측정하기 위함이다. 제안된 기법을 인위적으로 제시된 각 500개의 샘플을 가지는 6개의 독립신호와 1개의 종속신호를 대상으로 실험한 결과, 빠르고 정확한 변수의 선택이 이루어짐을 확인하였다.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.9 no.6
• /
• pp.657-664
• /
• 2014

This study is targeted at users of social networks to investigate motives, motives based on information exchange, network density, and reliability of information, recognizing network. Result motivation is a social network analysis of information-seeking motivation, social influence motivation, entertainment motivation, motivation network formation, respectively. Network density is also information seeking motivations, social influence motivation, entertainment showed differences in motivation, information about the reliability of the difference between the difference was in all the motivational factors.

• Korean Journal of Childcare and Education
• /
• v.8 no.5
• /
• pp.67-87
• /
• 2012

The purpose of this study is to investigate childcare teacher's knowledge of child safety education. In this study, Novak and Gowin(1984)'s concept map was applied to investigate knowledge of safety instruction. Forty teachers working at child centers were completed the concept map. The results of this study were as follows: First, Teachers used 194 higher concepts. 'Safe action in dander situation'(77.8%) was most frequently used higher concept and followed by 'traffic safety('35%). Second, Teachers used 1,270 dependent concepts. Third, there was significant difference in density according to frequency of safety instruction. These results can be meaningful to decide the contents of child safety instruction for teacher's education and as a basic data for teacher's education.

• Journal of Astronomy and Space Sciences
• /
• v.14 no.1
• /
• pp.24-32
• /
• 1997

We have calculated the velocity distribution of wind driven by Alfven waves. The assumed initial number density of wind can affect the line profiles because it produces the change in the velocity distribution under the mass conservation. Initial density $N_O=5.5{\times}10^{12}/cm^3$ is chosen for a proper initial density from the observation by Schroder(1986). The wind models for $N_O=10^9,10^{10},10^{11},5.5{\times}10^{12}/cm^3$ are calculated at ${phi}$=0.06 and ${phi}$=0.78. The line profiles for lower initial density show the strong emissions and narrow absorptions because of their steeper velocity gradients.