• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.260-261
• /
• 2010

본 논문에서는 제작된 콘텐츠에 배경음을 간편하고 효과적으로 혼합하기 위해서 녹음된 신호(전경음)를 분석하여 배경음 에너지를 적응적으로 조절하는 배경음혼합 알고리즘을 제안한다. 이를 위해, 제안된 알고리즘은 등청감 곡선 (equal-loudness curve) 및 linear prediction (LP) 분석에 기반하여 전경음신호의 청감 에너지 및 음성신호 존재여부를 결정한다. 이에 따라 전경음에 음성신호가 존재하는 경우에는 음성이 명확하게 들릴 수 있도록 혼합된 배경음의 에너지를 하향 조절하고, 반대로 전경음에 음성신호가 존재하지 않는 경우에는 배경음이 명확하게 들릴 수 있도록 혼합된 배경음의 에너지를 상향 조절한다. 제안된 알고리즘의 효율성을 검증하기 위해, 고정 가중치를 이용하여 배경음을 혼합하는 경우와의 음질 선호도 조사를 실시한 결과, 제안된 알고리즘에 대한 높은 선호도를 보였다.

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.2
• /
• pp.190-198
• /
• 2002

In general, we have used the Viterbi algorithm of Speech recognition for decoding. But a decoder in speaker verification has to recognize same word of every speaker differently. In this paper, we propose a noble decoding algorithm that could replace the typical Viterbi algorithm for the speaker verification system. We utilize for the proposed algorithm the speaker adaptation algorithms that transform feature vectors into the region of the client' characteristics in the speech recognition. There are many adaptation algorithms, but we take MLLR (Maximum Likelihood Linear Regression) and MAP (Maximum A-Posterior) adaptation algorithms for proposed algorithm. We could achieve improvement of performance about 30% of EER (Equal Error Rate) using proposed algorithm instead of the typical Viterbi algorithm.

• The Journal of the Acoustical Society of Korea
• /
• v.19 no.3
• /
• pp.45-50
• /
• 2000

This paper proposes a delta energy method based on Parameter Filtering(PF), which is a speech segmentation algorithm for text dependent speaker verification system over telephone line. Our parametric filter bank adopts a variable bandwidth along with a fixed center frequency. Comparing with other methods, the proposed method turns out very robust to channel noise and background noise. Using this method, we segment an utterance into consecutive subword units, and make models using each subword nit. In terms of EER, the speaker verification system based on whole word model represents 6.1%, whereas the speaker verification system based on subword model represents 4.0%, improving about 2% in EER.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.11b
• /
• pp.565-568
• /
• 2001

We have fabricated $Bi_{3.25}La_{0.75}Ti_{3}O_{12}(BLT)$ thin(200nm) films on the $Pt/Ti/SiO_{2}/Si$ substrates using a MOD(Metalorganic decomposition) method with annealing temperature from $550^{\circ}C$ to $750^{\circ}C$. The structural properties of the films examined by x-ray diffraction. The layered-perovskite phase obtained above $600^{\circ}C$. Scanning electron micrographs showed uniform surface composed of rodlike grains. The grain size increased with increasing annealing temperature. The BLT thin films showed little polarization fatigue test up to $3.5{\times}10^{9}$ bipolar cycling at a 5V and 100kHz.

• Annual Conference on Human and Language Technology
• /
• 1992.10a
• /
• pp.423-433
• /
• 1992

낱자 또는 음절들을 매우 짧게 제시하고 차폐로 지운 후 두 낱자 중하나를 선택하는 과제 (Reicher, 1969)를 사용하여 음절 맥락에서 낱자 지각에 영향을 주는 변수들과 그 배후 지각과정들을 검토하였다. 자모 조합유형이 음절 내 낱자의 정체 파악에 체계적으로 영향을 주고 있었고, 예상과는 달리 음절 내의 낱자들이 단독으로 제시되는 낱자 보다 잘 파악되지 않는 음절열등효과가 일관되게 관찰되었다. 본 연구의 세 실험 결과들을 상호작용활성화 모형과 초보지각자-기억자 모형에 비추어 논의하였다.

• Proceedings of the Korean Institute of Electrical and Electronic Material Engineers Conference
• /
• 2001.11a
• /
• pp.565-568
• /
• 2001

We have fabricated $Bi_{3.25}$$La_{0.75}$ti$_3$O$_{12}$ (BLT) thin(200nm) films on the Pt/Ti/SiO$_2$/Si substrates using a MOD(Metalorganic decomposition) method with annealing temperature from 55$0^{\circ}C$ to 75$0^{\circ}C$. The structural properties of the films examined by x-ray diffraction. The layered-perovskite phase obtained above $600^{\circ}C$. Scanning electron micrographs showed uniform surface composed of rodlike grains. The grain size increased with increasing annealing temperature. The BLT thin films showed little polarization fatigue test up to 3.5x10$^{9}$ bipolar cycling at a 5V and 100kHz.kHz.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2011.04a
• /
• pp.433-436
• /
• 2011

The effect of plasma on flow characteristics in subsonic flow in backward-facing step is studied. The velocty of main flows are 0.5 m/s. DBD plasma is using for flow control. Stainless foil and polymide films are used as an electrode and dielectric material. The change of flow characteristics are shown by different of plasma generation region in fluid flow.

• Tuberculosis and Respiratory Diseases
• /
• v.42 no.3
• /
• pp.413-418
• /
• 1995

Carcinosarcoma is an uncommon pulmonary malignancy characterized by carcinomatous parenchyma and sarcomatous stroma. Pulmonary carcinosarcoma represented about 1% of the resected lung tumors. The predominant clinical features are productive cough, chest pain, dyspnea, hemoptysis, bronchiectasis, and atelectasis, but alternate atelectasis is rare. We report a case of pulmonary carcinosarcoma associated with alternate atelectasis of the right upper and lower lobe.

• Journal of Labour Economics
• /
• v.43 no.2
• /
• pp.109-133
• /
• 2020

This paper estimates a matching function in Korea by using the method of Borowczyk-Martins et al. (2013) in order to mitigate endogeneity problems. The endogeneity was controlled through the GMM estimation by approximating the unobserved matching efficiency with the ARMA(p,q) process. The monthly data from the Economically Active Population Survey and Labor Force Survey at Establishments from June 2009 to December 2019 were used for the estimation. The matching elasticity that indicates the elasticity of new hires with respect to vacancies was estimated to be 0.859. When excluding 2019 samples, which had a large number of direct jobs provided by the government, the matching elasticity was lowered to 0.755, but still higher than that of other countries.

• Proceedings of the KIEE Conference
• /
• 1999.11c
• /
• pp.757-759
• /
• 1999

여러 사람의 얼굴들 중에 특정 개인 얼굴을 찾는 문제나 인식하는 문제는 최근들어 법 집행이나 상업적 목적 등 여러 응용분야에서 요구되고 있어 학문적으로 활발히 연구되어지고 있다. 얼굴 인식 기술은 여러 가지 방법으로 연구되어 왔는데 그 중 'PCA를 이용한 얼굴 인식 방법'이 가장 효율적인 방법으로 알려졌다. 그러나 이 방법은 조명 변화에 따라 정확성이 떨어지는 단점이 있다. 그래서 Histogram equalization을 이용해 조명 변화에 영향을 줄였다. 그리고 인식의 정확성을 유지하면서 eigenface를 추출하는데 시간을 줄이기 위해 웨이블렛 변환을 이용해 저주파 성분이 포함된 영역만을 추출, 그 부분을 입력영상으로 사용해 입력 영상에서 처리해야하는 차원을 줄여 특징 추출하는데 시간을 감소시켰다. 그 결과 특징 추출하는데 시간을 크게 줄어든 반면, 심한 조명 변화에서도 90%이상의 높은 인식률을 유지할 수 있었다.