• Speech Sciences
• /
• v.15 no.4
• /
• pp.85-96
• /
• 2008

Automatic speech recognition has severe performance degradation under noisy environments. To cope with the noise problem, many methods have been proposed. Most of them focused on noise-robust features or model adaptation. However, researchers have overlooked utterance verification (UV) under noisy environments. In this paper we discuss UV problems based on the normalized confidence measure. First, we show that UV performance is also degraded in noisy environments with the experiments of an isolated word recognition. Then we observe how the degradation of UV performances is suffered. Based on the UV experiments we propose a modeling method of the statistics of phone confidences using sigmoid functions. For obtaining the parameters of the sigmoidal models, the particle swarm optimization (PSO) is adopted. The proposed method improves 20% rejection performance. Our experimental results show that the PSO-NCM can apply noise speech recognition successfully.

• Proceedings of the Korean Society for Technology of Plasticity Conference
• /
• 1998.06b
• /
• pp.1-7
• /
• 1998

본 연구에서는 3차원 솔리드 모델러로부터 Modeling한 후 제작된 model 에 사용자가 원하는 측정 위치를 지정함으로써 자동적으로 측정용 NC Data를 얻을 수 있는 알고리즘을 설계하고 이를 Program화하였다. 3-D Modeler로서 Unigraphics를 사용하였으며, 공작기계는 FANUC Controller를 장착한 Machining Center이다. 측정용 센서는 접촉식 Probe를 사용하였다. 본 연구를 통하여 CAD시스템에서 구축된 Geomaatric Model로부터 직접 측정을 위한 NC data를 얻고 이 data를 가공후의 측정 Data와 비교하여 봄으로써 제품의 가공정밀도를 확인해 볼 수 있으며, 가공시의 간섭 check, 잔삭가공여유, 과절삭등 다양한 정보를 얻을 수 있을 것이다. 향수 지속적인 연구개발을 통하여 data의 통계처리 및 공작기계의 오차 보정등을 고려하여 줌으로써 기계상에서의 측정 data에 대한 신뢰도를 구축할 수 있을 것으로 사료된다.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.676-679
• /
• 2006

The media transport in automatic office machines such as printers, ATMs, copying machines is achieved by a complicated belt system. The system generally uses a crowning roller and belt which has been well-known for its intrinsic belt centering advantage during its operation. Since the modern office machines require precise high operating speed, stabilization of media transporting system has been one of the important issues of the machine design. Even a minor defect of the belt or the roller in the transport system directly affects its operating stability. This paper delivers a simulation technique that combines a multi-body dynamics analysis routine and a FEM based flexible continuum modeling for the efficient simulation of the flexible media transport problems.

• Transactions of the Korean Society of Automotive Engineers
• /
• v.25 no.2
• /
• pp.149-156
• /
• 2017

Various kinds of accessories(e.g., clock, radio, automatic door locks, alarm devices, etc.) or unit components (e.g., black box, navigation system, alarm, private audio, etc.) require dark current even when the vehicle power is turned off. However, accessories or unit components can be the causes of excessive dark current generation. It results in battery discharge and the vehicle's failure to start. Therefore, immediate detection of abnormal dark current and response are very important for a successful repair job. In this paper, we can increase the maintenance efficiency by presenting a standardized diagnostic process for the measurement of the dark current and the existing problem. As a result of the absence of a system to block the dark current in a vehicle, diagnosis and repair were performed immediately by using a standardized dark current diagnostic process.

• Proceedings of the Korean Information Science Society Conference
• /
• 2006.06c
• /
• pp.205-207
• /
• 2006

유비쿼터스 컴퓨팅 환경의 실현 가능성이 높아지면서 Automatic identification and Data Capture(AIDC)에 대한 많은 연구들이 이루어지고 있다. AIDC 기술 중 하나인 Radio Frequency IDentification(RFID)는 라디오 주파수에 따라 반응하는 태그를 이용하여 사람이나 사물을 인식하는 기술이다. 또한 RFID 특성을 지원하면서 발생 이벤트를 효율적으로 처리하기 위해 다양한 미들웨어가 개발되고 있다. 그러나 미들웨어의 지원에도 불구하고 RFID 시스템 개발자는 많은 부가적 지식을 알아야만 한다. 이 점을 해결하여 RFID 기술을 편리하게 사용하도록 돕는 Business Aware Framework(BizAF)이 제시되었다. 본 논문에서는 BizAF를 기반으로 여러 애플리케이션을 개발한 경험을 바탕으로 RFID 비즈니스 이벤트들을 생성하기 위한 모델링 프로세스를 제시하고자 한다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.2 no.3
• /
• pp.40-49
• /
• 1992

In this paper, an approach to identify fuzzy rules is proposed. The decision of the optimal number of fuzzy rule is made by means of fuzzy c-means clustering. The identification of the parameters of fuzzy implications is carried out by use of genetic algorithms. For the efficinet and fast parameter identification, the reduction thechnique of search areas of genetica algorithms is proposed. The feasibility of the proposed approach is evaluated through the identification of the fuzzy model to describe an input-output relation of Gas Furnace. Despite the simplicity of the propsed apprach the accuracy of the identified fuzzy model of gas furnace is superior as compared with that of other fuzzy modles.

• Journal of the Korea Society for Simulation
• /
• v.11 no.3
• /
• pp.13-22
• /
• 2002

This paper describes an automated evaluation of electrostatic field for micro motors whose sizes range 10 to 103um. Electric field modeling in micro motors has been generally restricted to in-plane two-dimensional finite element analysis (FEA). In this paper, the actual three-dimensional geometry of the micro motor is considered. An automatic FE mesh generation technique, which is based on the fuzzy knowledge processing and computational geometry techniques, is incorporated in the system, together with one of commercial FE analysis codes and one of commercial solid modelers. The system allows a geometry model of concern to be automatically converted to different FE models, depending on physical phenomena to be analyzed, electrostatic analysis and stress analysis and so on. The FE models are then exported to the FE analysis code, and then analyses are peformed. Then, analytical analysis and FE analysis about the torque generated by electrostatic micro motor are performed. The starting torque is proportional to $V^2$, the calculated starting torque from the two-dimensional analytical solutions are three times larger than those from the three-dimensional FE solutions.

• Phonetics and Speech Sciences
• /
• v.6 no.2
• /
• pp.9-20
• /
• 2014

In this paper, we propose an automatic fluency evaluation algorithm for English speaking tests. In the proposed algorithm, acoustic features are extracted from an input spoken utterance and then fluency score is computed by using support vector regression (SVR). We estimate the parameters of feature modeling and SVR using the speech signals and the corresponding scores by human raters. From the correlation analysis results, it is shown that speech rate, articulation rate, and mean length of runs are best for fluency evaluation. Experimental results show that the correlation between the human score and the SVR score is 0.87 for 3 speaking tests, which suggests the possibility of the proposed algorithm as a secondary fluency evaluation tool.

• Computers and Concrete
• /
• v.9 no.4
• /
• pp.293-310
• /
• 2012

This paper discusses a new constitutive model called the high-rate brittle microplane (HRBM) model and also presents the details of a new software package called the Virtual Materials Laboratory (VML). The VML software package was developed to address the challenges of fitting complex material models such as the HRBM model to material property test data and to study the behavior of those models under a wide variety of stress- and strain-paths. VML employs Continuous Evolutionary Algorithms (CEA) in conjunction with gradient search methods to create automatic fitting algorithms to determine constitutive model parameters. The VML code is used to fit the new HRBM model to a well-characterized conventional strength concrete called WES5000. Finally, the ability of the new HRBM model to provide high-fidelity simulations of material property experiments is demonstrated by comparing HRBM simulations to laboratory material property data.

• Geophysics and Geophysical Exploration
• /
• v.5 no.2
• /
• pp.71-77
• /
• 2002

We developed a modeling and visualization software that can analyze 3-dimensional seismic data. The software divides 3 dimensional space into a series of vertical and horizontal polygons, and allows the various seismic attributes and other spatial information to be stored on these polygons. The program can pick a particular pattern in semi-automatic mode, and store the pattern in the spatial DB. The pattern can be modeled and visualized in 3 dimensional space.