• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.345-348
• /
• 2002

음성인식 기술은 크게 음성인식과 화자인식 기술의 두 가지로 분류된다. 현재는 음성인식 기술이 널리 연구되고 있지만 점차 화자인식 기술의 중요성이 대두되고 있다. 본 논문에서는 화자인식 기술의 한 가지 분류로 임의 화자를 식별하기 위한 화자식별 기술을 연구 대상으로 하고 있으며, 신경회로망을 이용한 화자식별 시스템의 특징 추출 방법을 제시하고 그에 따른 성능을 비교하고 있다. 식별 단계에서 26명의 78개의 음성 샘플을 신경회로망의 역전파 알고리듬을 이용하여 학습하고, 테스트용으로 한 화자의 음성샘플이 사용되어 식별된다. 신경회로망의 입력 변수는 특징 파라미터로 선형예측계수, Mel-주파수 켑스트럼계수와 웨이블릿을 이용한 켑스트럼 계수를 사용하였다. 그 결과로써 화자식별 시스템의 신경회로망 모델2의 입력으로 혼합된 특징 파라미터를 사용한 경우가 다른 파라미터들을 사용한 경우와 비교하여 8.46~21.53%의 차를 가지고 가장 좋은 성능을 나타내었다.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.10 no.5
• /
• pp.449-458
• /
• 2000

In this paper, we propose a feed identification method using neural network to predict feed in crude distillation unit. The proposed FINN(feed identifier by neural network) is functionally composed of two modes-training mode and prediction mode. Also, we implement a neural network-based soft sensor system using Borland C++(3.0) Builder. The effectiveness of the proposed neural network-based feed identification method is shown by simulation results.

• Proceedings of the Korean Operations and Management Science Society Conference
• /
• 1993.04a
• /
• pp.275-281
• /
• 1993

본 발표에서는 신경회로망을 이용하여 연속음성중에서 키워드를 인식하는 방법을 설명한다. 연속음성에서 파형소편 및 음절을 식별하는 휴리스틱 알고리즘을 개발하였고, 연속음성을 음절단위로 파형소편 스펙트럼분석(선형예측법)으로 특성치를 추출하였다. 음절의 특성치는 코호넨 신경회로망을 통하여 학습을 시켰으며, 연속음성중 키워드인식은 먼저 음절을 인식하여 단어를 찾고, 인식된 단어가 키워드와 일치하는가를 확인한다. 본 연구의 의의는 파형소편 및 음절식별 알고리즘을 통하여, 크기불변성(Scaling invariance), 시간불변성(Time warping 및 Time-shift invariance), 중복성제거의 문제점을 해결하였고, 신경회로망의 학습을 통하여 화자독립적인 연속음성인식시스템 구축의 기반을 확립한데 있다. 본 음성인식모델은 학교구내 전화번호 안내시스템으로 활용단계에 있으며 전화번호뿐만아니라 주소안내시스템으로도 활용될 예정이다. 또한 자동차 운전보조시스템 및 주행안내시스템의 음성명령에 응용될 수 있는데, 예로 음성명령은 "핸들 좌로 20도", "시청까지 주행", "시청 지도안내"등이 될 수 있다. 현재 자동차 운전보조시스템은 컴퓨터 화면상 모의동작시스템으로 운영되고 있다. 본 음성인식모델은 화자종속시 90%이상, 화자독립시 70%의 인식결과를 보였다.시 90%이상, 화자독립시 70%의 인식결과를 보였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.215-218
• /
• 1997

A dynamic back-propagation neural network is addressed for adaptive neural control system to approximate non-linear control system rather than static networks. It has the capability to represent the approximation of nonlinear system without mathematical analysis and to carry out the on-line learning algorithm for real time application. The simulated results show fast tracking capability and adaptive response by using dynamic back-propagation neurons.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.2E
• /
• pp.57-64
• /
• 1995

In this study, we examine the applicability of the classifier based on an artifical neural network (ANN) for the low-frequency acoustic signal in shallow water environment. The estimations of the Doppler shift and frequency spreading effect at 220 Hz reveal the frequency variation of less than 2 Hz with time This small variation enables the ANN-based classifier to identify signals using only tonal frequency information. The ANN consists of 4 layers, and has 60 input processing elements (PEs) and 4 output PEs, respectively. When measured tonal signals in the frequency 200-250 Hz are applied to the ANN-based classifier, the classifier can identify more than 67% of the signals for instantaneous frame and more than 91% for averaged one over 5 frames.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.35 no.10
• /
• pp.912-917
• /
• 2007

In this paper system identification of a small UAV via neural networks is tried and the estimated parameters are then compared to those obtained by Fourier Transform Regression and Maximum Likelihood Estimation Techniques. With the estimated parameters a linear system is constructed and simulated to compare to the flight data. The results show that parameter identification using neural networks is comparable to the existing techniques

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.38 no.5
• /
• pp.14-22
• /
• 2001

This paper presents identification and control designs using neural networks for a class of multivariable nonlinear servomechanisms. A proposed neuro-controller is a combination of linear controllers and a neural network, and is trained by indirect neuro-control scheme. The proposed neuro-controller is implemented and tested on an IBM PC-based two 2 bar systems holding an object, and is applicable to many de-motor-driven precision multivariable nonlinear servomechanisms. The ideas, algorithm, and experimental results arc described. Moreover, experimental results are shown to be superior to those of conventional control.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.1
• /
• pp.106-113
• /
• 2019

In this paper, we propose an algorithm to classify the received active pulse when the active sonar system is operated as a non-cooperative mode. The proposed algorithm uses CNN (Convolutional Neural Networks) which shows good performance in various fields. As an input of CNN, time frequency analysis data which performs STFT (Short Time Fourier Transform) of the received signal is used. The CNN used in this paper consists of two convolution and pulling layers. We designed a database based neural network and a pulse feature based neural network according to the output layer design. To verify the performance of the algorithm, the data of 3110 CW (Continuous Wave) pulses and LFM (Linear Frequency Modulated) pulses received from the actual ocean were processed to construct training data and test data. As a result of simulation, the database based neural network showed 99.9 % accuracy and the feature based neural network showed about 96 % accuracy when allowing 2 pixel error.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.6 no.3
• /
• pp.487-492
• /
• 2002

In this paper, we proposed a control method of an unknown nonlinear system using a dynamical neural network. The proposed method is composed of neural network of state space model type, performs for a unknown nonlinear system, identification with using the dynamical neural network, and then a nonlinear adaptive controller is designed with these identified informations. In order to verify the effectiveness of the proposed method, we simulated one-link manipulator. The simulation results showed the effectiveness of using the dynamical neural network in the adaptive control of one-link manipulator.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.494-497
• /
• 2002

In this paper, we proposed a control method of an unknown nonlinear system using a dynamical neural network. The proposed method performs for a nonlinear system with unknown system, identification with using the dynamical neural network, and then a nonlinear adaptive controller is designed with these identified informations. In order to verify the effectiveness of the proposed method, we simulated one-link manipulator. The simulation results showed the effectiveness of using the dynamical neural network in the adaptive control of one-link manipulator.