• Proceedings of the IEEK Conference
• /
• 2000.09a
• /
• pp.815-818
• /
• 2000

본 논문에서는 감정 표현 음성 합성 시스템을 구현하기 위해서, 감정 음성 데이터베이스의 음향학적 특징인 피치, 에너지, 지속시간, 스펙트럼 포락에 대해 분석한 결과와 문법적 요소인 품사에 따른 감정 음성 데이터의 피치 변화를 분석하였다. 분석 결과, 기본 주파수, 에너지, 지속시간, 스펙트럼 포락은 감정 표현에 중요한 영향을 미치는 것으로 나타났으며, 전반적으로 화남과 기쁨의 감정이 평상과 슬픔의 감정 보다 피치 및 에너지의 변화가 크게 나타났으며, 특히 기쁜 감정의 경우 부사, 관형사, 연결어미, 조사, 접미사에서 피치 변화가 많았으며, 화난 감정의 경우, 관형사, 명사, 용언, 접미사에서 피치 변화가 높게 나타났다. 이러한 분석 결과를 적용해 감정 음성을 합성하기 위하여, 평상 음성에 각 감정 음성의 운율 요소를 적용하여 감정 음성을 합성하여 평가한 결과, 기쁜 감정은 기본 주파수의 변화에 의해 86.7%, 화난 감정은 에너지의 변화에 의해 91%, 슬픈 감정은 음절지속시간의 변화에 의해 76.7%가 각각 올바른 감정으로 인지되었다.

• Digital Contents
• /
• no.12 s.103
• /
• pp.44-45
• /
• 2001

최근 디지털화의 가속, 통신 인프라의 확충 등으로 전자, 정보 통신 기기들이 단일 네트워크로 연결되어 영상 및 음향정보를 서로 공유할 수 있는 환경이 마련되고 있다. 다시 말해 가정 내 모든 기기들이 네트워크화되어 마치 하나의 기기처럼 작동이 가능하게 되는 홈네트워킹의 구축이 가능해 지고 있는 것이다. 최근 LG경제 연구소에서 발행한 한 보고서에 의하면 홈네트워킹 관련 기술의 급속한 발전으로 새로운 사업 기회가 등장하고 있다. 이를 기회로 적극 활용하기 위해서는 국내 기업들의 기술력 제고, 전략적 제휴 확대 등을 통한 전반적인 경쟁력 강화가 시급하다고 지적하고 있다.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.05a
• /
• pp.286-286
• /
• 2004

초고속 정보통신망 구현에 필수적인 광학 연결용 부품인 광 Ferrule은 광섬유 커넥터의 중요부품으로서 우수한 치수정밀도와 내마모성을 갖추어야 한다. 현재, Ferrule로 사용되는 재료는 지르코니아, 알루미나 등 세라믹 물질이 많이 사용되고 있으며 기계적 강도나 내마모성 면에서는 우수하지만, 피연마속도가 석영섬유에 비해 현저히 작아서 특수한 연마방법을 채택해야 하며 성형성이나 가공성이 나빠 생산효율이 낮고 비용이 비싸다는 단점을 가지고 있다.(중략)

• Journal of KIISE:Software and Applications
• /
• v.28 no.10
• /
• pp.699-705
• /
• 2001

As it became possible to generate 3D sound on a PC environment due to the advances in computing performance and digital signal processing technology, 3D sound technology gains its focus in the multimedia area Specifically a two-channel based 3D sound technology is being studied by many researchers because of its space efficiency and economical structure. While the positional sound effect is simple in its implementation, the moving sound effect has many problems to be resolved as there are only discrete measured point of HRTF database. In this paper, we propose the method of generating smooth moving sound in a two-channel based 3D sound technique with respect to generating smooth trajectory, and the interpolation method of discrete measured HRTF data. We perform the tests in the PC environment and prove the utility of the proposed method.

• The Journal of the Acoustical Society of Korea
• /
• v.39 no.1
• /
• pp.24-31
• /
• 2020

In this paper, the deep convolutional neural network for sound event detection is studied. Especially, the end-to-end neural network, which generates the detection results from the input audio waveform, is studied for weakly supervised problem that includes weakly-labeled and unlabeled dataset. The proposed system is based on the network structure that consists of deeply-stacked 1-dimensional convolutional neural networks, and enhanced by the skip connection and gating mechanism. Additionally, the proposed system is enhanced by the sound event detection and post processings, and the training step using the mean-teacher model is added to deal with the weakly supervised data. The proposed system was evaluated by the Detection and Classification of Acoustic Scenes and Events (DCASE) 2019 Task 4 dataset, and the result shows that the proposed system has F₁-scores of 54 % (segment-based) and 32 % (event-based).

• The Journal of the Acoustical Society of Korea
• /
• v.21 no.8
• /
• pp.766-771
• /
• 2002

In this study, we have developed a 3-dimensional diagnostic ultrasonic sector probe using a convex type ultrasonic probe with 128 active elements. The probe was made to operate at the center frequency of 4.5㎒ with the bandwidth of 66%. The driving part was designed to rotate the axis of the convex probe by means of a step motor equipped with reduction gears and spur gears attached to the motor so that the probe could enable us to acquire a series of 2-dimensional images to construct a 3-dimensional image. Acoustic cover of the probe was made of polymers to have the same radius of rotation as that of the convex probe. The controllability of the rotation angle and the structural stability of the probe were confirmed through experimental 3-dimensional images obtained using the developed 3-dimensional sector probe.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.29 no.2
• /
• pp.209-217
• /
• 2011

This study incorporate hydroacoustic sampling for bathymetry and sediment survey in Won Cheon reservoir, Suwon city, Korea. Bathymetric and sedimentation surveys were conducted using a echo sounder system and subbottom profiler in the reservoirs. Data were collected using echo sounder systems and subbottom profiler linked to a GPS, to maximize data accuracy and vessel use, and geo-referenced using a DGPS enabling the acoustic data to be used in a GIS. Echo sounder and subbottom survey data were merged within geographic information system(GIS) software to provide detailed visualization and analyses of current depths, pre-impoundment topography, distribution, thickness, and volume estimates of lacustrine sediment, and water storage capacity. These data and analyses are, necessary for development of long term management plans for these reservoirs and their watersheds.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.4
• /
• pp.46-63
• /
• 1995

This thesis described an unlimited vocabulary connected speech recognition system using Time Delay Neural Network(TDNN). The recognition unit is the diphone unit which includes the transition section of two phonemes, and the number of diphone unit is 329. The recognition processing of korean connected speech is composed by three part; the feature extraction section of the input speech signal, the diphone recognition processing and post-processing. In the feature extraction section, the extraction of diphone interval in input speech signal is carried and then the feature vectors of 16th filter-bank coefficients are calculated for each frame in the diphone interval. The diphone recognition processing is comprised by the three stage hierachical structure and is carried using 30 Time Delay Neural Networks. particularly, the structure of TDNN is changed so as to increase the recognition rate. The post-processing section, mis-recognized diphone strings are corrected using the probability of phoneme transition and the probability o phoneme confusion and then the eojeols (Korean word or phrase) are formed by combining the recognized diphones.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.2
• /
• pp.32-39
• /
• 1999

In this paper, the characteristics of the ray tracing method (RTM) based on the cylindrical wave are discussed for the high frequency vibration analysis of two-dimensional structures. A ray tube describing the emanating cylindrical wave is used to derive the governing equation for incident reflected, and transmitted ray tubes which satisfies the condition at the coupled boundary. The suggested ray model is applied to panel array structures, and the predicted results for 2-panel, 3-panel, and 4-panel array structures are compared to those by Statistical energy analysis (SEA) and Wave intensity analysis(WIA). More enhanced prediction was obtained compared to the SEA, and similar prediction performance was observed to the WIA. Additionally, the RTM has a novel feature that it can estimate the spatially smoothed distribution of vibration energy and vibration intensity. It is expected that the present RTM can be used as one of the useful tools for the high frequency vibration analysis of two-dimensional coupled structures.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.303-307
• /
• 1998

본 논문에서는 고주파수 대역의 진동 전달 현상을 묘사할 수 있는 원형파를 기본 개념으로 한 레이 추적기법(RTM)을 개발하였다. 2차원 원형파를 묘사할 수 있는 레이튜브 개념을 제안하고, 진동 연성경계에서의 입사파, 전달파, 그리고 반사파가 경계조건을 만족하도록 관계식을 유도하였다. 이를 평판 연속 구조물이 고주파수 진동 전달 해석에 적용하여 보았으며, 유사 해석 기법인 SEA 및 파동 인텐시티해석법(WIA)과의 비교를 수행하여 보았다. 그 결과, SEA에 비하여 이론해와 더 잘 맞는 예측 결과를 관찰할 수 있었으며 WIA와 유사한 결과를 보임을 확인할 수 있었다. 본 논문에서 제안된 RTM은 진동전도해석(VCA)과 같이 지역적 정보를 알 수 있는 장점도 가지고 있으며, 2차원 진동구조물의 고주파수 진동 전달 해석시 기존이 SEA 및 WIA를 대체할 수 있는 기법으로 판단된다.