• 한국멀티미디어학회논문지
• /
• 제9권1호
• /
• pp.23-32
• /
• 2006

본 논문에서는 언어모델과 음향모델을 개선함으로써 단위 숫자음의 인식성능 최적화에 대해 설명한다. 언어모델은 한국어 단위 숫자음 문장의 문법적 특징을 분석하고, Finile State Network(FSN) 노드를 두 음절로 구성하여 오 인식률을 감소시켰다. 음향모델은 단 음절로 구성되어 발성기간이 짧고 조음이 많이 생기는 불명확한 음소, 음절의 분할로 인한 오 인식을 줄이기 위해 인식단위를 반음절 쌍으로 하였다. 인식단위의 특징을 효과적으로 모델링하기 위해 특징부분에서 K-means 알고리즘으로 군집화 하여, 상태를 분할하는 변형된 연쇄 상태 분할방법을 이용하였다. 실험 결과 제안된 언어모델의 적용 후 동일 문맥종속 음소모델에서 10.5%, 음향모델에서 인식단위를 반음절 쌍으로 하였을 경우 문맥종속 음소모델에 비해 12.5%, 변형된 연쇄 상태분할을 하였을 경우 1.5%의 인식률을 향상시킬 수 있었다.

• 한국음향학회지
• /
• 제13권1호
• /
• pp.71-83
• /
• 1994

본 논문은 한국어 문어면환(TTS : Text-to-Speech) 시스템내에서의 음성합성시 음질 및 자연성 개선을 위한 연구 결과이다. 합성방법으로는 단음절단위의 파형을 시간영역에서 분석(표1)하여 규칙합성에 필요한 매개변수(표2)를 추출하여 규칙합성시켰다. 실험에 사용된 음절은 한국어 발음 대사전의 빈도순위에 따라 V형 19개, CV형 80개, VC형 30개, CVC형 100개등 총 229음절을 선정하여 규칙합성시켰다. 규칙합성음의 평가방법으로는 229개의 규칙합성음중 음절 유형별로 15개씩 무작위로 추출한 합성음을 사전지식이 없는 임의의 그룹을 선정하여 이해도, 명료도, 잡음감, 자연성등 4가지 항목에 대하여 주관적인 오피니온 평가를 수행하였다. 실험결과, 합성음의 음질은 대단히 명료한 수준이었으며, 운율요소의 제어결과는 지속시간(장단)과 악센트(강약)의 제어(그림 9, 그림 10)가 가능하였으며, 피치주기(억양)의 제어도 Lagrange 보간법을 사용함으로써 가능하였다(그림 11, 그림 12).

• 한국음향학회지
• /
• 제6권2호
• /
• pp.38-47
• /
• 1987

한국어는 발음상의 특징과 구조에 의해서 음소철로 분리가 가능하므로, 한국어를 자음 음소, 모음 음소, 받침 음소로 나눌 수 있다. 분리된 각각의 음소들을 편자기 상관계수를 이용하여 분석하였는데, 이 때 예측 차수는 15차이다. 분석 실험에서 동일한 음소들은 그 특성이 거의 유사하였다. 한국어 단음 675개를 자음 음소. 모음 음소, 받침 음소로 각각 분리하여 인식한 결과 각각 $85.0(\%)$, $90.7(\%)$, $85.5(\%)$의 인식률을 얻었고, 이 음소들을 결합시킨 단음에서는 $72.1(\%)$의 인식률을 얻었다. 따라서, 이와 같은 방법을 이용하여 한국어 단음을 작은 데이터 양으로 처리 시간을 단축시킬 수 있고, 더 나아가 한국어의 모든 단음, 단어, 문장 둥을 인식할 수 있다.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 2000년도 7월 학술대회지
• /
• pp.166-177
• /
• 2000

The purpose of this study was to create letters for of nonliterary Pulang tribe in Thailand those who immigrant from China. illiterate Pulang tribe hand down their tradition by primary oral culture therefore their tradition can't initiate and keep, moreover, it may disappear throughout history. So it is expected to crusade against unlettered people. The scheme of research adopted in this study was a minority race who habitate at the northern Machan, Chiangrai in Thailand. It is not only analysis of language but also the eradication of literacy and the research based on linguistic, ethnolinguistic, and primary oral culture. Five Pulang people who live in that area were chosen for creating letters. By using the I. P. A., after each word was listen to their pronunciation one by one it was described and repeated this process several times; the material words and humanbody were pointed in front of them while other words were described by gesture. For final description, number of people were in the lineup for listening the sound of words and phrases to sentences. In the first stage, it was an analysis segmental of Pulang: vocoid, contoid and diphthong were described with each sample syllables and words. The suprasegmental were studied with intonation and juncture of the words in the second stage. Two words were compared and different meanings within their intonation and juncture were shown. At the end of this part, each case of phonemic or morphophonemics representation described the juncture in the words. In the third stage, minimal pairs were analyzed with vowels and consonants and described in free variation based on words. In the last stage, syllable structure in open syllable and closed syllable was studied and then each syllable of its structure was analyzed with samples. There were thirty-two phonemes in apong Pulang as follows: seven vocoids; a, i, e, o, u, ${\ae}$, and $\wedge$, one diphthong; wu, 24 contoids; b, c, d, f, g, h, j, k, k, 1, m, n, ${\eta}, {\;}p^{h}$, p, p, r, s, s, sh, t, t, w, and y. Their pronunciations of p, s, d, $p^{h}$, j, and t are frequently used in speech and are unique in triphthong. Moreover, most of the words used initial and final consonant cluster.

• 대한예방한의학회지
• /
• 제17권1호
• /
• pp.163-179
• /
• 2013

Objectives : After analyzing the proficiency of medical communication of the students in College of Korean Traditional Medicine using standardized patients, we suggests ways to improve clinical practice in the future class and medical communication curriculum development. Methods : 20 students before clinical practice class (3rd grade) and 20 students after 1 year clinical practice class (4th grade) participated and did their medical interview on Standardized patient. They were evaluated on patient-physician communication skills by standardized patients and professor evaluator. In addition to be evaluated on patient-physician relationship, medical interview skills by professor evaluator. Results : As follows in the evaluation of clinical practice with standardized patients 1. More than half of the participated students regardless of their grade received poor score in their medical communication evaluated by SP(Standardized patient) and PE(Professor evaluator). 2. Greeting, History taking parts were higher in the 4th students who received 1 year clinical practice class, but verbal-nonverbal response, voice tone parts were higher in the 3rd students who do not received clinical practice lesson. 3. Pronunciation&Voice tone parts were higher in the male students but, gathering information part was higher in the female students. Conclusions : We think that the current clinical practice lessons are insufficient as a way to learn and improve medical knowledge and medical communication skills, and it is necessary a new form of clinical practice class. Participatory lesson using standardized patient could be a good alternative of that in the future class.

• 한국지역지리학회지
• /
• 제17권4호
• /
• pp.381-394
• /
• 2011

한국내 외국인 인구가 약 126만 명으로 현재 우리사회는 다인종 다문화사회로 급속하게 변화하고 있다. 유, 초, 중, 고에 취학하고 있는 다문화 가정 학생수도 약 3만6천명 (2010년 기준)이 넘고 있어 앞으로 다문화가정의 학생이 더욱 증가할 것으로 예상할 수 있는데 실제로 국제결혼과 이민자들의 가장 큰 고민거리중의 하나는 자녀들에 대한 편견과 차별에 대한 우려이다. 연구의 결과는 다문화 가정 학생들이 피부색, 언어의 발음, 학교생활에 부적응 등으로 집단 따돌림을 당하고 있는 실정에 있다. 다문화교육 대상학년은 사실상 초등학교 l 학년부터 지도되어야 하고 다문화 가정 학생과 일반학생이 통합 교육으로 지도해야하고, 다문화 교육 주요자료는 영상자료, 사진자료를 사용하는 것이 좋을 것으로 밝혀졌다. 안내 자료는 책자나 CD자료가 적합하다고 하였고 지도 시기는 교과와 관련하여 지도하는 것이 바람직하다고 하였다.

• 한국치위생학회지
• /
• 제15권1호
• /
• pp.73-79
• /
• 2015

Objectives: The purpose of this study was to investigate the factors influencing on the denture satisfaction in the low income elderly people. Methods: The subjects were 143 elderly people from 60 to 75 years old wearing dentures and receiving consistent follow-up in the public health center in Busan. A self-reported questionnaire was filled out by the elderly people from February 1 to March 1, 2013. The questionnaire consisted of denture satisfaction, social variables, and psychological variables. Data were analyzed by ANOVA and multiple regression analysis using SPSS 18.0. The sociodemographic characteristics included gender, age, and chronic diseases. The recognition of oral health included pronunciation, denture maintenance, mastication ability, and education for denture care. The questionnaire was measured by Likert 5 scale. Results: Gender, chronic disease, denture care instructions, and the self-preception of the oral health after denture treatment were closely correlated with denture satisfaction(p<0.001). Age and the number of repairs were very important factor to denture satisfaction(p<0.05). Elderly women were more satisfied with denture than men and those who had no chronic diseases tended to be more satisfied with denture. Those who received oral care instructions were more satisfied with the denture than those who did not. The younger age group and no repairing prosthetic group tended to be more satisfied with the denture. Conclusions: It is important to provide the denture management services to the low income elderly when they demand the services. The national dental health policy must be focused on connection of the elderly people denture services with the public health center.

• 복식
• /
• 제56권6호
• /
• pp.72-86
• /
• 2006

The study aimed to materialize the meaning of the names through analyzing names that have synonyms and ambiguous words. The subjects of the study are seven foreign language study books such as $\ll$譯語類解 Yeokeoyuhae $\gg$ $\ll$譯語類解補編 Yeokeoyuhae - supplementary book$\gg$, $\ll$同文類解 Dongmunyuhae$\ggl$, $\ll$象語類解 Mongeoyuhae$\gg$, $\ll$象語類解補編 Mongeoyuhae- supplementary book$\gg$, $\ll$俊語類解 Oaeeoyuhae$\gg$ and $\ll$方름類釋 Bangeonyuseok$\gg$. The study selected vocabularies related with headgear, socks, clothing from those vocabulary books. The results are as follows: 1. The result focusing on the same Chinese terms in multiple forms has shown that '帽子(Moja), 斗逢(Dubong), 頭面(Dumyen), 襪(Mal), 木?(Mokgeuk), 　據子(Nanmalja), 樓子(Oja), 小樓子(Sooja), 養衣(Saui), 團領(Danlyeng), 掛子(Goija), 齊肩掛(Jegyengae), 彩兒(Sama), 單補(Dango)' have differences in Korean annotations, and it has been found that these differences are spelt as synonyms in general. 2. The result focusing on the same Korean terms in multiple forms has shown that Gamto, Gat, Satgat, Bolkki, Gatmo, Neoneol, Gatdongot, Ddambatgijeoksam, Nuyeok, Tyeollik, Goui, Baji, Gatdongot, Hatot, Dongdoljji, Yusam, Chima and ete. have 2 or more Chinese spellings. Among these, Neoneol, Ddambatgijeoksam, Nuyeok, Tyeollik, and etc. seem to have used similar Chinese spellings, and Gat, Dongdoljji, and etc. seem to have used Chinese spellings according to the multiple forms in each country. 3. There have also been records of terms seldom used at present such as Dalogi, Neoneol, and etc., and it has been found that Mahulae, Doljji, Dunja, and etc. are the terms have been geographically influenced by the northern race as they have been left as a dialect in North Korea currently. 4. Mahulae, Kulimae, Neolku, and etc. may be inferred as they have originated from Manchu language through pronunciation of each country spelt in translated books.

• 한국음향학회지
• /
• 제39권5호
• /
• pp.483-489
• /
• 2020

본 논문에서는 저자원 환경의 음성인식에서 음향 모델의 성능을 높이기 위한 음향 모델 학습 방법을 제안한다. 저자원 환경이란, 음향 모델에서 100시간 미만의 학습 자료를 사용한 환경을 말한다. 저자원 환경의 음성인식에서는 음향 모델이 유사한 발음들을 잘 구분하지 못하는 문제가 발생한다. 예를 들면, 파열음 /d/와 /t/, 파열음 /g/와 /k/, 파찰음 /z/와 /ch/ 등의 발음은 저자원 환경에서 잘 구분하지 못한다. 자기 주의 메커니즘은 깊은 신경망 모델로부터 출력된 벡터에 대해 가중치를 부여하며, 이를 통해 저자원 환경에서 발생할 수 있는 유사한 발음 오류 문제를 해결한다. 음향 모델에서 좋은 성능을 보이는 Time Delay Neural Network(TDNN)과 Output gate Projected Gated Recurrent Unit(OPGRU)의 혼합 모델에 자기 주의 기반 학습 방법을 적용했을 때, 51.6 h 분량의 학습 자료를 사용한 한국어 음향 모델에 대하여 단어 오류율 기준 5.98 %의 성능을 보여 기존 기술 대비 0.74 %의 절대적 성능 개선을 보였다.

• 한국조리학회지
• /
• 제4권
• /
• pp.211-224
• /
• 1998

This study is research about coinage and interpretation of Japanese Menus in luxury hotels in Seoul. Japanese Menus consist of three categories. These can be expressed A+B+C groupings. A represents Ingredients, C represents the cooking method, while B shows the name of a region(B1), the seasoning(B2), and a word that expresses the shape of the food(B3), etc. B can be flexible in meaning. However, the setting, A+B+C, is not always used strictly for these category meanings. Ingredients, A, is sometimes used independently(ex, ぃくとろろ), and at other times B1 + C, B2 + C or B3 + C are used. Sometimes A+C is used without B. The most general expression is Ingredients + the method for cooking(A+C Type). By knowing the menu description the food type and cooking method can be discovered. Most concrete method is Ingredients + procedure for seasoning + cooking method(A+B2+C Type) show how something is made and what kind of seasoning. This method is frequently used for roasted dishes with seasoning. Food which is expressed by A, Ingredients, does not require a complicated cooking process and is fixed by a conventional cooking process without any explanation ; delicacies(珍味), an hors d'oeuvres(前菜), sushi(Japanese vinegared rice delicacies) and sliced raw fish. There are two obstacles in interpreting Japanese Menus. First, we cannot look up the menu words easily in a dictionary because the mixing of Chinese Characters and the pronunciation of them differs from Korean chinese characters. Secondly, the names of Japanese food are inserted with a name of a place or another symbolic word, so they sometimes cannot be translated. We should offer various and accurate information for foreign customers because various Non-Japanese people use these restaurants frequently. This will enable them to enjoy themselves more comfortably. Therefore, you should decide the words carefully and provide an explanation about the complicated parts of the Menu when you work with Menu copywriter.