• 대한음성학회지:말소리
• /
• 제66호
• /
• pp.21-39
• /
• 2008

The purpose of this study is to analyze the intonational slope characteristics of the Korean news utterances. Prosodic phrases were analyzed in terms of the K-ToBI labeling system. In addition, the change of intonation contour that occurs throughout the sentences was discussed in terms of types of media and gender. Results showed that the overall declination of the intonation contour of radio and male revealed a gentler slope than that of TV and female, respectively. While the regression of the top line slope showed male's higher $R^2$ with the number of words, the base line slope of the radio and female was proved to be highly influenced from the number of syllables, words, and prosodic phrases. A lot more independent variables statistically affected to the base line slope. This means that the base line slope was strongly related to the variables, the top line slope, otherwise, could be more freely fluctuated due to the light correlation with them.

• The Journal of the Acoustical Society of Korea
• /
• 제17권2E호
• /
• pp.26-30
• /
• 1998

In this paper, we compose neural network predictive HMM(NNPHMM) to provide the dynamic feature of the speech pattern for the HMM. The NNPHMM is the hybrid network of neura network and the HMM. The NNPHMM trained to predict the future vector, varies each time. It is used instead of the mean vector in the HMM. In the experiment, we compared the recognition abilities of the one hundred Korean syllables according to the variation of hidden layer, state number and prediction orders of the NNPHMM. The hidden layer of NNPHMM increased from 10 dimensions to 30 dimensions, the state number increased from 4 to 6 and the prediction orders increased from 10 dimensions to 30 dimension, the state number increased from 4 to 6 and the prediction orders increased from the second oder to the fourth order. The NNPHMM in the experiment is composed of multi-layer perceptron with one hidden layer and CMHMM. As a result of the experiment, the case of prediction order is the second, the average recognition rate increased 3.5% when the state number is changed from 4 to 5. The case of prediction order is the third, the recognition rate increased 4.0%, and the case of prediction order is fourth, the recognition rate increased 3.2%. But the recognition rate decreased when the state number is changed from 5 to 6.

• 한국정보처리학회논문지
• /
• 제7권4호
• /
• pp.1122-1131
• /
• 2000

The efficiency of thesytem depends upon an accurate extraction capability of index terms in the system of information search or in that of automatic index. Therefore, extraction of accurate index terms is of utmost importance. This report presents the generation methods of composition noun for efficient index term extraction by using words of high frequency appearance, so that the right documents can be found during information search. For the sake of presentation of this method, index terms of composition noun shall be extracted by applying the rule of composition and disintegration to the nouns with high frequency of appearance in the documents, such as those with upper 30%∼40% of frequency ratio. In addition, for he purpose of effecting an inspection of validity in relation to a composition of high frequency nouns such as those with upper 30∼40% of frequency ratio as presented in this report, it proposes an adequate frquency ratio during noun composition. Based upon the proposed application, in this short documents with less than 300 syllables, low frequency omissions were noticed, when composed with nouns in the upper 30% of frequency ratio; whereas the documents with more than 30 syllables, when composed with nouns in he upper 40% of frequency ration, had a considerable reduction of low frequency omissions. Thus, total number of index terms has decreased to 57.7% of these existing and an accurate extraction of index terms with an 85.6% adequacy ratio became possible.

• 아동학회지
• /
• 제34권1호
• /
• pp.71-86
• /
• 2013

The purpose of this study was (1) to examine the level of learning disabilities reflected in the written expression and writing performance of at-risk children with ADHD, (2) to investigate the level of differences in writing learning disabilities and writing performance depending on ADHD subtypes, and (3) to explore the error types and contents in the written expression of at-risk children with ADHD. The participants in this study were 46 upper grade elementary school children. They were firstly screened by teacher nomination, and only participants with a K-ARS score of over 17 were then selected to be among the 46 children involved in this study. Two further tests were then carried out : K-LDES as an index of learning disabilities in written expression and BASA-writing as an index of writing performance. The results showed that the at-risk children with ADHD possibly had comorbid writing learning disabilities. They were significantly different in terms of the number of total syllables, errors and correct syllables that they produced, in comparison to normal children. But there were no differences as regards the level of learning disabilities in terms of written expression and writing performance based on ADHD subtypes. As regards the implications of these results for future research, we suggested that there is a need for the identification of comorbid writing learning disabilities in ADHD assessment.

• 대한안전경영과학회지
• /
• 제13권3호
• /
• pp.29-35
• /
• 2011

Safety and health related information for the proper use and handling of pesticides is usually printed on the surface of the pesticide products (bottle type or bag type) in the form of texts. But, the guidelines or standards for the appropriate presentation of the texts for the pesticide products are most vague or not practical. Thus, this study aimed to provide the preliminary guidelines for the text sizes based on the legibility experiments. Total twenty subjects from two age groups (young: n=10, old: n=10, five males and five females in each group) participated in the experiment. First, subjects read the text cards presented in the distance of 50cm from the eyes of the subjects. Eight different text card sets were prepared for different font type(thick gothic-type and fine gothic-type), thickness of font(plain and bold), and number of syllables (2 and 3 syllables). When subjects read the cards, the correctness of reading (correct or wrong) was recorded and the degree of discomfort (from 1: no discomfort at all to 4: can't read at all) was also evaluated for all the text sizes. Results showed that the character size should be 4 pt or larger for the young subjects to read at least one word correctly in all the text conditions. For the old subjects to read at least one word correctly, the character size should be five pt or larder. The average of the minimum character sizes for 100% correct answer is 6.1 pt for young subjects and 10.5 pt for old subjects, respectively.

• 패션비즈니스
• /
• 제14권6호
• /
• pp.11-22
• /
• 2010

The main purpose of this research was to examine the linguistic characteristics of domestic men's formal wear brand name. Four linguistic characteristics of language type, combined structure type of language, word class, length of brand name were investigated in this research and also examined the difference between brand type. For sample selection, the 209 men's fashion brands were selected from '2009 Korea Fashion Yearbook' and then, 25 brands which could not collect proper informations about the brand name or naming were excluded. Among total 184 men's brand names, 66 men's formal wear brands were selected and studied. For data analysis, quantitative evaluation of the frequency and qualitative evaluation have been used. The result as follows.; (1) Seven language types were found in domestic men's formal wear brand names. English has been used the most, then followed by Italian and French. (2) For combined structure type of brand name language, the single word used the most, followed by separately combined word type, artificially combined word, and unified word type. (3) The most frequently used the type of word class was noun, and followed by phrase, adjective, and verb. In the noun type, 6 different types which expressed a person, concrete & abstract entity, place, acronym, and neologic were found. For phrase, only noun type was appeared, however, 6 out of 20 phrases were abbreviated type. All eight adjective brand names implied an attributive character of the brand such as 'Dainty' or 'Solus(Solo)'. (4) The long name used most and then followed by normal and short length of brand name. Looking by the number of syllable, 4 syllables appeared the most and then followed by 3, 5, 6, 2 & 7 showed the same rate, and 8 syllables. (5) The result which compared the difference according to each brand type showed a difference in its language type, language combined style, word class, but length of brand name.

• 음성과학
• /
• 제9권3호
• /
• pp.185-200
• /
• 2002

The current study confirmed the existence of an ip prosodic level in French intonation structure, as previously proposed by Sun-Ah Jun & $C\acute{e}cile$cile Fougeron (2000). However, in contrast to the previous suggestion of the plateau realized in an ip in several syntactic structures, the current study supposed that the plateau doesn't come from the different type of syntactic structures but arise from the unspecified syllables without any PA in an ip. Because if we limited ip phrasal tone to the syntactic structure, it would be difficult to find the more general reasons of ip level. Besides /Hi/ and /$H^*$/ we also used /$Hi^*$/ for the focused syllable in the current study. In emphasized sentences, in general, /$Hi^*$/ appeared in the first or second syllable of a leftward AP in an ip and /$H^*$/ in the final syllable of a rightmost AP of an ip, In contrast to these PAs, /$Hi^*$/ might appear in any syllable in an ip, but not to far from /$H^*$/ because the duration time and length t of plateau realized between /$Hi^*$/ and /$H^*$/ or /Hi/ and /$H^*$/ would make an essential harmonious rhythmic unit, Therefore, the current study determined the duration time and the number of syllables realized in each plateau in an ip level composed of more than one AP. As a phrase constituent structure, there is a practical need for intermediate prosodic units to allow for generalization over the many possible combinations of prosodic patterns that can occur. Further evidence is still needed to analyze and relate the different pitch ranges of the plateau of an ip according to the syntactic structure, to identify the considerable character in the French prosodic hierarchy.

• 대한음성학회:학술대회논문집
• /
• 대한음성학회 1996년도 10월 학술대회지
• /
• pp.456-463
• /
• 1996

In this paper, we discussed three cases to see the effects of the characteristics of Hangul writing system. In applications such as computer Hangul shorthands for ordinary people and pushbuttons with Hangul characters engraved, we found that there is much advantage in using Hangul. In case of Hangul Transliteration, we discussed some problems which are related with the characteristics of Hangul writing system. Shorthands use 3-set keyboards in England, America, and Korea. We saw how ordinary people can do computer Hangul shorthands, whereas only experts can do computer shorthands in other countries. Specifically, the facts that 1) Hangul characters are grouped into syllables (syllabic blocks) and that 2) there is already a 3-set Hangul keyboard for ordinary people allow ordinary people to do computer Hangul shorthands without taking special training as with English shorthands. This study was done by the author under the codename of 'Sejong 89'. In contrast like QWERTY or DVORAK, a 2-set Hangul keyboard cannot be used for shorthands. In case of English pushbuttons, one digit is associated with only one character. However, by engraving only syllable-initial characters on the phone pushbuttons, we can associate one Hangul "syllable" with one digit. Therefore, for a given number of digits, we can associate longer words or more meaningful words in Hangul than in English. We discussed the problems of the Hangul Transliteration system proposed by South Korea and suggested their solutions, if available. 1) We are incorrectly using the framework of transcription for transliteration. To solve the problem, the author suggests that a) we include all complex characters in the transliteration table, and that b) we specify syllable-initial and -final characters separately in the table. 2) The proposed system cannot represent independent characters and incomplete syllables. 3) The proposed system cannot distinguish between syllable-initial and -final characters.

• 한국융합학회논문지
• /
• 제12권3호
• /
• pp.1-7
• /
• 2021

인공신경망 기계번역(Neural Machine Translation, NMT)은 한정된 개수의 단어만을 번역에 이용하기 때문에 사전에 등록되지 않은 단어들이 입력으로 들어올 가능성이 있다. 이러한 Out of Vocabulary(OOV) 문제를 완화하고자 고안된 방법이 서브 워드 분절(Subword Tokenization)이며, 이는 문장을 단어보다 더 작은 서브 워드 단위로 분할하여 단어를 구성하는 방법론이다. 본 논문에서는 일반적인 서브 워드 분절 알고리즘들을 다루며, 나아가 한국어의 무한한 용언 활용을 잘 다룰 수 있는 사전을 만들기 위해 한국어의 음절 중 종성을 분리하여 서브 워드 분절을 학습하는 새로운 방법론을 제안한다. 실험결과 본 논문에서 제안하는 방법론이 기존의 서브 워드 분리 방법론보다 높은 성능을 거두었다.

• Journal of Ecology and Environment
• /
• 제29권3호
• /
• pp.219-225
• /
• 2006

노랑턱멧새의 테마송과 변이를 연구하기 위해서, 6개 도에 산재한 16개 활엽수림지대에서 번식하는 45개체의 노랑턱멧새 수컷에서 3,245개의 노랫소리를 녹음하였다. 노랫소리를 구성하는 주된 요소이며 대용문자가 부여된 640개 음절의 조합을 분석하여 164 테마송, 1,024 변이로 분류하였다. 수컷 개체는 각각 $1{\sim}6$ 테마송을 가졌고, 평균 3.5개 레퍼토리를 가지고 있었다. 수컷 개체 사이에서 노랫소리를 구성하는 음절의 순서와 배열이 완전히 동일한 테마송은 하나도 없었다. 수컷 개체의 노랫소리를 구성하고 있는 음절의 수는 $5{\sim}14$개(평균 9.4개)였다. 수컷은 하나의 테마송에 변이를 만들기 위해서 음절을 첨가, 삭제, 대체하여 레퍼토리 크기를 효과적으로 증가시켰다. 하나의 테마송이 갖는 변이의 수는 평균 5.1($1{\sim}31$)개였고, 개체 변이성은 노랫소리의 말단부의 구성 요소에서 가장 높았다. PCA 분석에서 공유하는 음절에 기초를 둔 요인 I과 고유한 음절에 기초를 둔 요인 II에 의해서 노랑턱멧새 16 개체군은 서로 다른 군집을 형성하였다. 공유한 음절의 수에 기초를 둔 유사성 측정은 지역과 일치하는 강한 형태를 나타냈다. 양방향 유사도를 분석한 결과 16 개체군은 UPGMA 군집 속으로 나누어졌으며, 지리적인 거리가 증가함에 따라 유사도는 감소하는 경향을 보였다.