http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
오늘 본 자료
허만탁,김재창,Huh, Man-Tak,Kim, Jae-Chang 한국음향학회 1997 韓國音響學會誌 Vol.16 No.1
본 논문에서는 스펙트럼 포락선을 이용하여 음성을 인식하기 위한 새로운 전처리 방법을 제안한다. 이는 확산필터뱅크를 사용하여 스펙트럼 포락선을 추출하는 새로운 방법이다. 확산필터뱅크의 분석대역을 몇 개의 작은 대역으로 나눔으로써 확산회수를 줄였으며 차분회수를 늘임으로써 선택도를 높였다. 이 결과, 총처리시간을 대폭 줄였으며 스펙트럼의 변별력을 증가시켰다. 컴퓨터 시뮬레이션을 통하여 간단한 인식 알고리듬으로 실제 음성의 단모음 인식 실험을 해본 결과 3%의 인식율을 얻음으로써 확산필터뱅크가 많은 주파수 성분을 가진 음성의 주파수 분석을 이용하는 음성인식에 대단히 유효하다는 것을 확인하였다. In this paper, a new pre-processing method for the recognition of single vowels by use of spectrum envelope is presented. We use new extraction method of a spectrum envelope using the diffusion filter bank. By dividing analysis band of a diffusion filter bank into subbands, we decreased the number of diffusion process. And, by increasing the number of difference, we got higher selectivity. As a result of them, we reduced the total processing time, and got higher enhancement of discrimination. By getting 88.3% of average recognition rate for single vowels of natural voice through computer simulation. We confirmed it to be useful for speech recognition which use spectrum analysis of the voice signal to have many frequency components.
A Study on On-line Recognition System of Korean Characters
최석,김길중,허만탁,이종혁,남기곤,윤태훈,김재창,이양성,Choi, Seok,Kim, Gil-Jung,Huh, Man-Tak,Lee, Jong-Hyeok,Nam, Ki-Gon,Yoon, Tae-Hoon,Kim, Jae-Chang,Lee, Ryang-Seong The Institute of Electronics and Information Engin 1993 전자공학회논문지-B Vol.b30 No.9
In this paper propose a Koaren character recognition system using a neural network is proposed. This system is a multilayer neural network based on the masking field model which consists of a input layer, four feature extraction layers which extracts type, direction, stroke, and connection features, and an output layer which gives us recognized character codes. First, 4x4 subpatterns of an NxN character pattern stored in the input buffer are applied into the feature extraction layers sequentially. Then, each of feature extraction layers extracts sequentially features such as type, direction, stroke, and connection, respectively. Type features for direction and connection are extracted by the type feature extraction layer, direction features for stroke by the direction feature extraction layer and stroke and connection features for stroke by the direction feature extraction layer and stroke and connection features for the recongnition of character by the stroke and the connection feature extractions layers, respectively. The stroke and connection features are saved in the sequential buffer layer sequentially and using these features the characters are recognized in the output layer. The recognition results of this system by tests with 8 single consonants and 6 single vowels are promising.
a-b-c 프레임에 근거한 AC-DC 변환기용 세가지 제어전략
이동수(Dong-Su Lee),김지웅(J-Woong Kim),전성즙(Seong-Jeub Jeon),허만탁(Man-Tak Huh) 전력전자학회 2005 전력전자학술대회 논문집 Vol.- No.-
Three control strategies that can be applied to a three-phase AC-DC converter under unbalanced input voltage condition are discussed. Analytic solutions are given on the a-b-c frame, and are very simple, direct and intuitive. All of the strategies guarantee sinusoidal input currents. All the control functions, including decoupled current control and PWM, are implemented on the a-b-c frame. Accordingly, the controller is very simple and robust. The proposal is verified by simulation and experiments on a prototype operating at 15 ㎑.
스위칭 펄스폭에 따른 SC의 등가저항 변화에 관한 연구
許萬鐸 釜山工業大學校 1981 論文集 Vol.22 No.2
The SC equivalent resistance which is variable with switching pulse width is described. For developing the theory, the conducting resistance of a switch is very useful which have been neglected ever. The theory is discussed and compared with measured data. The results of experiment agree with the theory.
許萬鐸,金康彦 釜慶大學校 1996 釜慶大學校 論文集 Vol.1 No.2
In this paper, we improved analyzing method of a spectrum envelope¹ using the diffusion filter bank.???? We increased the number of channel of filter bank. By dividing analysis band into subbands, we decreased the number of diffusion process. And, by increasing the number of difference, we got higher selectivity. As a result of them, we reduced the total processing time, and got higher enhancement of discrimination. Through computer simulation, we showed that average recognition rate for single vowels of real voice is 88.3[%]. Hence We confirmed it to be useful for speech recognition which use spectrum analysis to analyze frequency components.⁴
許萬鐸 釜山工業大學校 1995 論文集 Vol.37 No.-
In this paper, a diffusion filter bank which needs very less hardware than conventional one is introduced, and the results of extracting spectum from real speech signals by use of that are presented. The spectra extracted from the signals by simmulation have so smooth envelope that we can estimate it is suitable for speech recognition which demand many channels with diffrent center freqencies, and smooth spectrum of speech signals.