• Proceedings of the IEEK Conference
• /
• 1999.11a
• /
• pp.663-667
• /
• 1999

In this paper, we review the conventional speaker verification algorithm and present the text-dependent speaker verification system for application over telephone lines and its result of experiments. We apply blind-segmentation algorithm which segments speech into sub-word unit without linguistic information to the speaker verification system for training speaker model effectively with limited enrollment data. And the World-mode] that is created from PBW DB for score normalization is used. The experiments are presented in implemented system using database, which were constructed to simulate field test, and are shown 3.3% EER.

• Journal of the Institute of Convergence Signal Processing
• /
• v.12 no.4
• /
• pp.309-314
• /
• 2011

As a system becomes more complex, a design relies more heavily on a methodology based on high-level abstraction and functional verification. SystemVerilog includes characteristics of hardware design language and verification language in the form of extensions to the Verilog HDL. However, the OOP of System Veri log does not allow multiple inheritance. In this paper, we propose adoption of SystemC to introduce multiple inheritance. After being created, a SystemC unit is combined with a SystemVerilog-based verification environment using SystemVerilog DPI and ModelSim macro. Employing multiple inheritance of SystemC makes a design of a verification environment simple and easy through source code reuse. Moreover, a verification environment including SysemC unit has a benefit of reconfigurability due to OOP.

• Atmosphere
• /
• v.21 no.2
• /
• pp.197-208
• /
• 2011

The objective of this study is to describe the short-range forecast system of the Korea Air Force (KAF) and to verificate its performace in 2009 summer. The KAF weather prediction model system, based on the Weather Research and Forecasting (WRF) model (i.e., the KAF-WRF), is configured with a parent domain overs East Asia and two nested domains with the finest horizontal grid size of 2 km. Each domain covers the Korean peninsula and South Korea, respectively. The model is integrated for 84 hour 4 times a day with the initial and boundary conditions from National Centers for Environmental Prediction (NCEP) Global Forecast System (GFS) data. A quantitative verification system is constructed for the East Asia and Korean peninsula domains. Verification variables for the East Asia domain are 500 hPa temperature, wind and geopotential height fields, and the skill score is calculated using the difference between the analysis data from the NCEP GFS model and the forecast data of the KAF-WRF model results. Accuracy of precipitation for the Korean penisula domain is examined using the contingency table that is made of the KAF-WRF model results and the KMA (Korea Meteorological Administraion) AWS (Automatic Weather Station) data. Using the verification system, the operational model and parallel model with updated version of the WRF model and improved physics process are quantitatively evaluated for the 2009 summer. Over the East Aisa region, the parallel experimental model shows the better performance than the operation model. Errors of the experimental model in 500 hPa geopotential height near the Tibetan plateau are smaller than errors in the operational model. Over the Korean peninsula, verification of precipitation prediction skills shows that the performance of the operational model is better than that of the experimental one in simulating light precipitation. However, performance of experimental one is generally better than that of operational one, in prediction.

• MALSORI
• /
• no.67
• /
• pp.79-102
• /
• 2008

In this paper, we investigate the relation between Bayes factor and likelihood ratio test (LRT) approaches and apply the neighborhood information of Bayes factor to building an alternate hypothesis model of the LRT system. To consider the neighborhood approaches, we contemplate a distance measure between models and algorithms to be applied. We also evaluate several methods to improve performance of utterance verification using neighborhood information. Among these methods, the system which adopts anti-models built by collecting mixtures of neighborhood models obtains maximum error rate reduction of 17% compared to the baseline, linear and weighted combination of neighborhood models.

• Journal of the Institute of Convergence Signal Processing
• /
• v.10 no.4
• /
• pp.274-279
• /
• 2009

The flow of a universal system-level design methodology consists of system specification, system-level hardware/software partitioning, co-design, co-verification using virtual or physical prototype, and system integration. In this paper, verification environments based-on SystemVerilog and SystemC, one is native-code co-verification environment which makes prompt functional verification possible and another is SystemVerilog layered testbench which makes clock-level verification possible, are implemented. In native-code co-verification, HW and SW parts of SoC are respectively designed with SystemVerilog and SystemC after HW/SW partitioning using SystemC, then the functional interaction between HW and SW parts is carried out as one simulation process. SystemVerilog layered testbench is a verification environment including corner case test of DUT through the randomly generated test-vector. We adopt SystemC to design a component of verification environment which has multiple inheritance, and we combine SystemC design unit with the SystemVerilog layered testbench using SystemVerilog DPI and ModelSim macro. As multiple inheritance is useful for creating class types that combine the properties of two or more class types, the design of verification environment adopting SystemC in this paper can increase the code reusability.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3C
• /
• pp.257-267
• /
• 2009

This paper presents a training method for neural networks and the employment of MSE (mean scare error) values as the basis of a decision regarding the identity claim of a speaker in a recurrent neural networks based speaker verification system. Recurrent neural networks (RNNs) are employed to capture temporally dynamic characteristics of speech signal. In the process of supervised learning for RNNs, target outputs are automatically generated and the generated target outputs are made to represent the temporal variation of input speech sounds. To increase the capability of discriminating between the true speaker and an impostor, a discriminative training method for RNNs is presented. This paper shows the use and the effectiveness of the MSE value, which is obtained from the Euclidean distance between the target outputs and the outputs of networks for test speech sounds of a speaker, as the basis of speaker verification. In terms of equal error rates, results of experiments, which have been performed using the Korean speech database, show that the proposed speaker verification system exhibits better performance than a conventional hidden Markov model based speaker verification system.

• Proceedings of the Earthquake Engineering Society of Korea Conference
• /
• 2001.04a
• /
• pp.394-402
• /
• 2001

In order to lessen cable vibration, new cable damper system with high damping rubber was developed using the basis of the LRB design scheme. The analysis model of cable damper system incorporate voigt-kelvin damper model into the nonlinear cable analysis model. To achieve maximum damping capacity both reducing damper stiffness and developing high damping rubber were performed. As a result of verification test, the high damping rubber damper show its effectiveness in improving cable damping capacity.

• Proceedings of the IEEK Conference
• /
• 2003.07b
• /
• pp.939-942
• /
• 2003

Design verification problem is emerging as an important issue to detect any design errors at the early stage of the design. Conventionally, design verifications have been done using a simulation technique. However, this technique has been proved not to cover all potential design errors. Therefore, formal technique is often used to verify digital circuits as an alternative. In this paper we adopted formal verification technique and verified some important properties derived from our pipelined digital systems, using SMV (Symbolic Model Verifier). Our example shows that model checking method (one of formal verification techniques) can be effectively performed in verifying the large digital systems.

• KIISE Transactions on Computing Practices
• /
• v.23 no.4
• /
• pp.256-261
• /
• 2017

The notion of System of Systems (SoS), which is composed by many independent systems (i.e., Constituent Systems, CS), has emerged in various domains including social infrastructure. It is widely expected that complex requirements, which cannot be achieved in each CS-level, will be achieved in an SoS-level. While verification of SoS-level goal achievement is one of the most important problems, concrete case studies on SoS modeling and verification are still rare. In this paper, we focus on the fact that each CS performs an action for its own purpose by its own decision-making mechanism. We propose a novel Action-Benefit-Cost (ABC) SoS model which caters to the independent decision-making mechanisms of CSs. Using an abstract SoS example, this proposal provides a case study for the modeling and quantitative verification of the ABC SoS model.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.7
• /
• pp.1555-1562
• /
• 1995

This study represents the non-dimensional cutting force model. With the non-dimensional cutting force model it is possible to estimate efficiently the maximum cutting force during one revolution of cutter. Using the non-dimensional cutting force model, the feed rate and spindle speed are adjusted so as to satisfy the maximum cutting force and maximum machining error. To verify the accuracy and efficiency of the non-dimensional cutting force model, a series of experiments were conducted, and experimental results proved and verified the non-dimensional cutting force model. The NC toolpath verification system developed in this paper uses the non-dimensional cutting force model, so that it is effective for calculating the cutting force and adjusting the cutting conditions.