• Proceedings of the KIEE Conference
• /
• 2006.07a
• /
• pp.288-289
• /
• 2006

Conventional AGC controller implemented by considering ACE and bias factors always contains error terms due to inaccuracy of bias factors. To improve this steady state AGC performance, PI controller can be introduced to the ACE processing and plant controller part. Even though more rigorous inspections are required, a simple application shows that it can be very useful for the improvements of AGC performances.

• Phonetics and Speech Sciences
• /
• v.15 no.3
• /
• pp.75-82
• /
• 2023

Automatic speech recognition (ASR) has been revolutionized with deep learning-based approaches, among which self-supervised learning methods have proven to be particularly effective. In this study, we aim to enhance the performance of OpenAI's Whisper model, a multilingual ASR system on the Korean language. Whisper was pretrained on a large corpus (around 680,000 hours) of web speech data and has demonstrated strong recognition performance for major languages. However, it faces challenges in recognizing languages such as Korean, which is not major language while training. We address this issue by fine-tuning the Whisper model with an additional dataset comprising about 1,000 hours of Korean speech. We also compare its performance against a Transformer model that was trained from scratch using the same dataset. Our results indicate that fine-tuning the Whisper model significantly improved its Korean speech recognition capabilities in terms of character error rate (CER). Specifically, the performance improved with increasing model size. However, the Whisper model's performance on English deteriorated post fine-tuning, emphasizing the need for further research to develop robust multilingual models. Our study demonstrates the potential of utilizing a fine-tuned Whisper model for Korean ASR applications. Future work will focus on multilingual recognition and optimization for real-time inference.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.34 no.7
• /
• pp.45-54
• /
• 2006

General unmaned airship, in use of aviation photography, needs both airship -controller and camera-controller who work together in harmony. In oder to reduce this manpower and get the good Geographical Information Systems(GIS) data, it is necessary to use a autopilot controller which guides a exact path lines. This paper presents the autopilot control law base on classical PID control. Moreover, this paper shows the result of flight test, the procedure of gain tuning and LOS guidance algorism that is reduce a tracking error.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.613-618
• /
• 2022

효과적인 자연어 처리를 위해 제안된 Transformer 구조의 등장 이후, 이를 활용한 대규모 언어 모델이자 사전학습 모델인 BERT, GPT, OPT 등이 공개되었고, 이들을 한국어에 보다 특화한 KoBERT, KoGPT 등의 사전학습 모델이 공개되었다. 자연어 처리 모델의 확보를 위한 학습 자원이 늘어나고 있지만, 사전학습 모델을 각종 응용작업에 적용하기 위해서는 데이터 준비, 코드 작성, 파인 튜닝 및 저장과 같은 복잡한 절차를 수행해야 하며, 이는 다수의 응용 사용자에게 여전히 도전적인 과정으로, 올바른 결과를 도출하는 것은 쉽지 않다. 이러한 어려움을 완화시키고, 다양한 기계 학습 모델을 사용자 데이터에 보다 쉽게 적용할 수 있도록 AutoML으로 통칭되는 자동 하이퍼파라미터 탐색, 모델 구조 탐색 등의 기법이 고안되고 있다. 본 연구에서는 한국어 사전학습 모델과 한국어 텍스트 데이터를 사용한 자연어 처리 모델 산출 과정을 정형화 및 절차화하여, 궁극적으로 목표로 하는 예측 모델을 자동으로 산출하는 시스템의 설계를 소개한다.

• The Journal of the Korea Contents Association
• /
• v.10 no.4
• /
• pp.84-90
• /
• 2010

DBMS is used for a range of applications from data warehousing through on-line transaction processing. As a result of this demand, DBMS has continued to grow in terms of its size. This growth invokes the most important issue of manually tuning the performance of DBMS. The DBMS tuning should be adaptive to the type of the workload put upon it. But, identifying workloads in mixed database applications might be quite difficult. Therefore, a method is necessary for identifying workloads in the mixed database environment. In this paper, we propose a SVM workload classifier to automatically identify a DBMS workload. Database workloads are collected in TPC-C and TPC-W benchmark while changing the resource parameters. Parameters for SVM workload classifier, C and kernel parameter, were chosen experimentally. The experiments revealed that the accuracy of the proposed SVM workload classifier is about 9% higher than that of Decision tree, Naive Bayes, Multilayer perceptron and K-NN classifier.

• Journal of Convergence Society for SMB
• /
• v.4 no.1
• /
• pp.31-39
• /
• 2014

PID auto-tuning controller was designed via fuzzy logic. Typical values such as error and error derivative feedback were changed as heuristic expressions, and they determine PID gain through fuzzy logic and defuzzification process. Fuzzy procedure and PID controller design were considered separately, and they are combined and analyzed. Obtained auto-tuning PID controller by Fuzzy Logic showed the ability for less than 3rd order plant control. We also applied to reference tracking problem with the designed auto-tuning scheme.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2004.04a
• /
• pp.409-412
• /
• 2004

일반적으로 반도체 생산 라인으로부터 생산되는 제품의 수율은 반도체의 가격에 직접적인 영향을 미치는 인자로 고수율의 제품 생산을 위해서는 반도체 생산라인 내부의 공기를 청정하게 할 필요가 있으며 이를 위해 고성능의 필터들이 사용되고 있다. 이러한 필터의 재료는 초극세 섬유로 작업자의 부주의한 조작에 의해 필터 표면에 핀홀(pinhole)둥이 발생하기 쉽다. 이러한 핀홀은 육안으로 관측이 힘들뿐만 아니라 필터의 여과 성능에도 악영향을 미치는 요소로 필터 생산업체에서는 이의 검출 및 보수에 많은 인력이 투입되고 있다. 본 연구에서는 필터 표면에 발생된 핀홀을 검출할 수 있는 자동 테스트 장치를 제안함과 동시에 제안된 테스트 장치의 효율적 운영을 가능케 하는 최적 운전 파라미터를 퍼지 모델링 기법을 이용하여 튜닝하고자 한다.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2000.11a
• /
• pp.136-139
• /
• 2000

Controller refinement scheme to improve the performance of a conventional system automatically in frequency domain is proposed. The controller automatic tuning method features using experimental frequency responses of the conventional closed-loop system, the conventional controller, and the improved closed-Imp system; instead of poorly modeled plant due to non-linearities and disturbances. The improved closed-loop system characteristics is automatically acquired by the conventional closed-loop system characteristics and the proposed performance index in system bandwidth. And the proper controller is realized by least squares approximation in frequency domain. To testify the usefulness of the approach, experimental results of robot path-tracking control applied with various controllers is used, and then is analyzed with respect to a equivalent proportional controller. Experimental results and analytic results are well-matched.

• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.99-104
• /
• 2001

Discrete-time controller design is proposed using feedback system identification in frequency domain. System Stability imposed by a new controller is checked in the function of a conventional closed-loop system, instead of a poorly modeled plant due to non-linearity and disturbance as well as unstable components, etc. The stability of the system is evaluated in view of Popov criterion. All the equations are formulated in the framework of the discrete-time system. Simulation results are shown on the plant with input saturation components, DC disturbance and a pure integration.

• Proceedings of the KSME Conference
• /
• 2001.06b
• /
• pp.664-669
• /
• 2001

Continuous-time controller design is proposed using feedback system identification in frequency domain. System stability imposed by a new controller is checked in the function of a conventional closed-loop system, instead of a poorly modeled plant due to non-linearity and disturbance as well as unstable components, etc. The stability of the system is evaluated in view of Nyquist stability. All the equations are formulated in the framework of the discrete-time system. Simulation results are shown on the plant with input saturation and DC disturbance.