한글 받침 발음법의 페트리 네트 표현

이계영,임재걸 동국대학교 경주대학 1995 東國論集 Vol.14 No.-

Research on natural language and speech processing in terms of the man-machine interface is now in progress. On account of Korean speech idiosyncrasies, speech processing has been subject to a searching examination for the relation of word to phonetic value. This paper proposed the Petri net representation methods for the phenomenon of Korean vocalization. The differences of a proposed new Petri net from standard Petri net are as follows : First, Partition of input and output layer. Second, the transition whose input layer is the same as the current marking fires. And, The proposed Petri net for the phenomenon of vocal sound could also be used to solve problems of the conversion of symbolic language(Hanguel) into phonetic language(Korean) to illustrate vocal sound changes in a symbolic manner.

칼슘투여에 따른 운동선수의 혈중 지질함량 변화 연구

이계영,신말순 대진대학교 1994 大眞論叢 Vol.2 No.-

The study was carried to investigate the effects of a long term calcium upon blood lipid metabolism in athletes, The effects of calcium supplementation and hematological variables which are related to calcium metabolism. Subjects(N=30), Who were 30 males of athletes, were randomly assigned to two group(p1acebo group, P15, experimental group N=15. Each subject in experimental group was ingested a tablet containing 500mg of calcium three times a day for 8 weeks. Hematological variable were measured pre and post 2, 4, 6, 8 weeks after the tablets ingestion. The results through the statiscal analysis of blood components, tatal-cholesterol, HDL-C, LDL-C, Total-cholesterol/HDL-C, Triglyceride in lipid components were as follow; 1) Total cholesterol was decreased after 2, 4, 6, 8 weeks in Calcium group but there was no significantly decreased in placebo group. 2) LDL-C was significantly decreased in calcium group after 2 weeks and maintained through out the entire experimental period(P<0.05), but there was no significantly decreased in placebo group. There were significant differences between two group(calcium group, placebo group). 3) HDL-C was significantly increased in calcium group after 4 weeks and maintained through out the entire experimental period(P<0.05), while showed no significant change in placebo group. There were significant differences between two group(calcium group, placebo group). 4) TC/HDL-C was decreased after 2, 4, 6, 8 weeks in calcium group but there was no significantly decreased in placebo group. There were significant differences between two group(P<0.05). 5) Triglyceride showed not significantly reduce after calcium supplement, not difference between two group(P<0.05).

건설자재 수평 운반 로봇

이계영,한창수 제어로봇시스템학회 2009 제어로봇시스템학회 국내학술대회 논문집 Vol.2009 No.9

Construction industry is still considered as labor intensive industry although mechanized construction technologies are applied. In constructing building, the distribution of construction materials is a large portion of construction process. Besides forklift which is for unloading heavy-weighted materials from truck, the usage of human powered cart is high. In this research, we propose a new motorized system transporting construction materials.

女子卓球選手들의 體型과 體力間의 關係

李啓暎,韓泰龍 全北大學校附設 스포츠科學硏究所 1987 논문집 Vol.- No.4

The objectives of this study were to study on the somatotype and physical fitness of woman table tennis players. For this purpose the writer surveyed the Heath-Carter Anthropometric somatotype method and physical fitness. The total 26 female athletes were selected and measured. The conclusion of this study were as follows : 1. The correlation coefficient between Endomorphy and power, and Muscular endurance were revealed as the certainly correlation. 2. The correlation coefficient between mesomorphy and muscular strength was revealed as the high correlation, Agility, power turn out certainly correlation. 3. The correlation coefficient between ectomorphy and power was revealed as the certainly correlation.

정보 검색을 위한 Optimal Relevance Feedback 시스템에 관한 연구

李桂英,李太庚 동국대학교 2007 慶州大學論文集 Vol.5 No.-

한국어 음운 변동 처리 규칙의 설계 및 구현

이계영,Lee, Gye-Young 한국정보처리학회 1998 정보처리논문지 Vol.5 No.3

한국어는 음운 변동 현상이 매우 발달되어 있다는 특징을 갖고 있다. 따라서, 음성 인식율의 제고와 음성 합성음의 자연스러움을 향상시키기 위해서는 음운 변동 현상을 비롯한 한국어의 모든 특징에 대한 체계적인 연구가 있어야 한다. 본 논문은 한국어의 제 특징 중에서 음운 변동 현상을 효율적으로 처리할 수 있는 규칙을 설계하고 구현함으로써, 한국어 음성 인식과 합성에 효율적으로 이용될 수 있음을 보인다. 음운 변동 규칙의 설계를 위하여 사용된 규정은 한글 맞춤법 통일안의 표준 발음법(7장 30항)이며, 일차적으로 각 규정별로 설계된 규칙을 최종적으로 27개 그룹으로 정리된 종성별 규칙을 제시하였다. 본 연구에서 제안된 음운 변동 처리 시스템은 한번의 규칙 적용으로 음운 변도 dgus상이 신속히 처리되는 잇점이 있으며, 단어 및 용언의 어간에 접속되는 정보에 대한 처리 내용까지도 제안된 규칙내에 수용하였기 때문에 문장 단위의 음성 인식과 합성 시스템의 연구에 도움을 줄 수 있다. Korean language is to be characterized by the rich vocal sound variation. In order to increase the probability of vocal sound recognition and to provide a natural vocal sound synthesis, a systematic and thorough research into the characteristics of Korean language including its vocal sound changing rules is required. This paper addresses an effective way of vocal sound recognition and synthesis by providing the design and implementation of the Korean vocal sound variation rule. The regulation we followed for the design of the vocal sound variation rule is the Phonetic Standard(Section 30. Chapter 7) of the Korean Orthographic Standards. We have first factor out rules for each regulations, then grouped them into 27 groups for eaeh final-consonant. The Phonological Change Processing System suggested in the paper provides a fast processing ability for vocal sound variation by a single application of the rule. The contents of the process for information augmented to words or the stem of innected words are included in the rules. We believe that the Phonological Change Processing System will facilitate the vocal sound recognition and synthesis by the sentence. Also, this system may be referred as an example for similar research areas.

"정보 검색을 위한 Optimal Relevance Feedback 시스템에 관한 연구"

李桂英,李太庚 동국대학교 경주대학 1986 論文集 Vol.5 No.-

A Information Retrieval System is assumed in which the index transformation consists of mapping stored information and requests from the natural language to multidimensional property vectors in a abstracts space. The end product of the index transformation is, however, a vector in the concept space in which the weight of concept that is the i-th component of the vector is indicative of the frequency of each concept in the original source (ABSTRACTS). The formulation of the optimal query corresponding to a particular set of ABSTRACTS has no direct implication on operation of information retrieval, since the set of ABSTRACT in question is the object of the retrieval search. Here, this kind of circularity is suggested a strong analogy to feedback control, and the optimal method of selecting a subject area is employed through implementation, the retrieved results are showed by experimentation, analyzed.

한국어 음성 합성을 위한 음가 변환 테이블 생성

이계영,임재걸,Lee, Gye-Young,Yim, Jae-Geol 대한전자공학회 2001 電子工學會論文誌-CI (Computer and Information) Vol.38 No.5

문법적으로 정확한 한국어 음성을 합성하려면 표준어 규정의 '표준 발음법'을 준용해야 한다. 따라서 한국어 음가 합성 시스템에 사용되는 한글을 음성으로 변환하여 주는 규칙은 '표준 발음법'을 완전하게 반영하며 또한 무결해야 한다. 기존의 연구에서는 표준 발음법을 검증없이 적용하여 왔고, 표준 발음법 자체에 모순이 있는가의 여부에 대해서도 체계적인 분석을 위한 시도가 진부하였다. 본 논문에서는 한국어 음가 생성의 기본 규칙으로 준용할 표준 발음법을 페트리 넷으로 모델링하여 표준 발음법의 일관성을 검증하였다. 그리고, 음운 변동 현상을 여러 단계로 나누어 차례로 적용한다든지, 변동된 단어에 대하여 처음부터 다시 변환 작업을 재수행하는 기존의 음가 생성 방법의 문제점을 해결하기 위하여 한번의 테이블 탐색으로 모든 음운 변동이 완료되는 한국어 음성 합성을 위한 음가 변환 테이블을 구현하였다. In order to synthesize grammatically correct Korean voices, we have to refer to the 'Standard Pronunciation Rules(SPR)' stated in the 'Standard Grammar of Korean Language.' Therefore, the rules that is used for a Korean-voice-synthesis system to find Korean voices corresponding to a given Korean sentence must completely reflect the SPR and must be sound. However, in the field of computer science they have just used the SPR without proving the completeness and soundness of their rules. In this paper, we construct a Petri net model for each rule of SPR, integrate all the Petri net models to build one big Petri net completely representing SPR, and analyse the Petri net to prove the consistency of it. Then, we transfer the Petri net model into a look-up table for Korean voice. Using this table, we can avoid the drawbacks of existing approaches such as going through several stages or repetitively applying a converting process.

음절 특징과 필터 뱅크를 이용한 음성 단어 인식에 관한 연구

이계영 동국대학교 경주대학 1990 論文集 Vol.9 No.-

This paper describes the DataBase design and hardware implemenation for Korean Word Recognition. the parameter of Korean Speech uses Formant, and the Korean Word Recognition is processed by the syllable synthesize method after the recognition of syllable units. The Hardware is implemented by the Filter Bank method. The strong point of this method has the possibility of a smaller data and real time processing about a large Word recognition. The result shows the word recognition of about 95.3% for 10 Korean words.

하악골 전돌증 환자의 악교정수술 후 교합평면에 관한 연구

이계영,장현중,이상한,김철환 대한악안면성형재건외과학회 2003 Maxillofacial Plastic Reconstructive Surgery Vol.25 No.5