• Annual Conference on Human and Language Technology
• /
• 2003.10d
• /
• pp.33-39
• /
• 2003

한국어에서는 복합 명사 생성이 매우 자유스럽다. 즉, 독립된 명사를 연속으로 붙여 쓰는 것이 가능하다. 하지만, 기계번역이나 정보 검색과 같이 복합 명사를 처리하는 시스템에서 정확한 분석을 위해서는 복합 명사를 다시 단일 명사들로 분해하는 과정이 필요하다. 본 논문에서는 한국어 복합 명사 분해를 위해 GECORAM(GEneralized Combination of Rule-based learning And Memory-based learning) 알고리듬을 제시한다. 규칙 학습 알고리듬의 장점은 생성된 학습 결과를 사람이 쉽게 이해할 수 있다는 점이지만, 다른 지도학습 알고리듬에 비해 성능이 떨어진다는 단점이 있다. 본 논문에서는 이를 위해 규칙 학습 알고리듬과 기억기반 학습을 결합하는 방법을 제시한다. 실험 결과, GECORAM 알고리듬은 규칙 기반 학습이나 기억 기반 학습을 단독으로 쓰는 경우보다 높은 정확도를 보였다.

• Speech Sciences
• /
• v.11 no.1
• /
• pp.65-72
• /
• 2004

The force of articulation is different between voiced and voiceless consonants in the binary opposition system. However, the Korean voiceless stop consonants have a triple opposition system: lenis, aspirated, and glottalized. The aim of this study is to find the primary distinctive feature between the force of articulation and the aspiration for the three different types of Korean stops. Two native speakers of the Seoul dialect participated to this study. The corpus was composed of less than eight syllabic words containing consonants in word-initial position and intervocalic position. Radiocinematography and Mingography were used to analyze the articulatory tension and acoustic characteristics. Korean stops have independent features of articulatory tension and aspiration, in which the indices are different according to position. However, in this system which does not have the opposition of sonority, the force of articulation is the primary distinctive feature and the feature of aspiration is subsidiary.

• Speech Sciences
• /
• v.11 no.3
• /
• pp.181-192
• /
• 2004

The metrical structures in orally produced the poetry were generally analyzed by accent, metre and syllable. The purpose of this study is to investigate of metrical structures of Korean modem poetry using computer implemented speech analysis system. Two famous poet's poems confidential talk, Miloe and 'A buddhist dance, Sungmu' were selected for prosodic analysis. The informant is 60 years old professor in major of Korean and French poetry. The syllable structures of poems were analyzed primarily by vowel timbers, which can classified compact and diffuse vowels according to the distance of F2-F1. The perception cues of consonants were analyzed by VOT and tensity features of articulation. Rhythm is classified by dactyl, anapest, trochee, spondee and iambic. As a result, syllable structures of Korean modem poetry were mainly CV and CVC and the reading times of each lines were 3-4sec for 12 and 15 syllables. Main metre of Korean modem poems constructed the Imbic and Anapest. The break of each lines were demarcated by grammatical structure or meaning rather than phonetic structures.

• Korean Journal of Cleft Lip And Palate
• /
• v.5 no.1
• /
• pp.1-9
• /
• 2002

Some children with Cleft Palate have shown a speech disorders after repaired surgical operation. A diagnostic evaluation of speech in children with cleft palates is important in preventing speech disorders. However, standard speech evaluation form for children with cleft palates has not yet developed in Korea. The purpose of this study is to make the standard speech evaluation form for children with cleft palates. Thirty control children group and ten children with cleft palate participated in this experiment. The test words are composed of meaningless two syllabic words containing the three different types of korean stop consonants,

• 한국어정보학회:학술대회논문집
• /
• 2017.10a
• /
• pp.330-332
• /
• 2017

개체명 인식(Named Entity Recognition, 이하 NER)은 인명(PS), 기관명(OG), 장소(LC), 날짜(DT), 시간(TI) 등에 해당하는 개체명에 일정한 태깅 값을 주어 그 정보를 가시화하는 작업이다. 한국어 개체명 인식은 아직 그 자질이 충분히 밝혀져 있지 않아 자연어 처리 분야의 발전을 더디게 하는 한 요소로 작용하고 있다. 한국어가 음절 기반으로 단어를 형성하고 비교적 어순이 자유롭다는 특성이 있기에, 이런 특징을 잘 포착할 수 있는 "음절 기반의 Convolutional Neural Network(CNN)"의 아키텍쳐를 제안하여 66.80%의 성능을 보였다. 이 방법을 사용하면 형태소 분석등 개체명 이전 단계에서 발생하는 오류에 의해 개체명 인식(NER)의 성능이 떨어지는 문제를 해결할 수 있고, 조사나 어미 등을 제거하기 위한 후처리를 생략할 수 있다.

• Speech Sciences
• /
• v.12 no.2
• /
• pp.7-16
• /
• 2005

The purpose, this experimental study was to identify pronunciation difficulties between male and female Korean speakers in their articulation of English vowels. Ten English mono-syllabic words were spoken six times by six male and six female college students. Formant frequencies were measured from sound spectrograms made by Pitch Works. Results show that Korean female speakers uttered English vowels more similar to those uttered by English native speakers than did Korean male speakers. While Korean male speakers could not readily distinguish between /i/ and /I/, /u/ and /v/, and /$\epsilon$/ and /ae/, respectively, Korean female speakers had difficulty only with /$\epsilon$/ and /ae/. The tentative results suggests that on the whole Korean speakers have difficulty in discriminating tense vowels from lax vowels, and they also have articulatory problems pronouncing low and back vowels such as /ae/. /a/ and /c/.

• Annual Conference on Human and Language Technology
• /
• 1994.11a
• /
• pp.288-294
• /
• 1994

훈민정음 해례에 따르면 한글문자는 음소 및 음절 문자 특성을 가지고 있다. 이러한 특성들을 컴퓨터 시스템에서 구현함에 있어서 야기되어 왔던 각종 문제를 분석한 다음 이들 문제들에 대하여 한글문자의 특성을 제약함이 없이 컴퓨터에 대한 기술을 개발함으로써 해결책을 모색한다. 본 논문은 훈민정음 해례에서 밝힌 한글 문자의 음소 및 음절 문자 특성에 따라서 기존의 코드 체계를 평가하며, 그리고 이들에 대한 구현 방안을 제시하고자 한다. 또한 이러한 특성을 반영한 한글 입출력들인 '셔블'을 개발하고 이에 대한 검증을 시도하였다.

• Phonetics and Speech Sciences
• /
• v.6 no.4
• /
• pp.109-115
• /
• 2014

Fricated nuclei in Nuosu Yi were found to be more correctly described as fricated vowels, rather than syllabic fricatives due to the presence of clear formant structures typical of front vowels. In this exploratory study, two types of fricated nuclei were examined: retroflex "yr" and non-retroflex "y". The retroflex nucleus "yr" had higher F1 and lower F3 than non-retroflex "y", indicating a lower tongue height. On the other hand, F2 was found to correlate not with nucleus retroflexion, but instead with onset consonant retroflexion: F2 was higher following retroflex onsets, in both vowels. This effect was persistent through the entire vowel, suggesting a phonological effect, rather than a coarticulatory one. Interpretation of the F2 results require accompanying articulatory data since the usual coupling of F2 and tongue backness does not always hold for retroflex vowels. Examining the articulation of the fricated nuclei in Nuosu Yi is a direction for future research.

• Speech Sciences
• /
• v.13 no.2
• /
• pp.35-46
• /
• 2006

This study investigated whether there are differences between Korean-English bilingual and Korean monolingual children on phonological awareness skills. Participants were 11 Korean-English bilingual children and 12 Korean monolingual children. The children's ages ranged between 6 and 7 years. The results were as follows. First, the bilingual children significantly outperformed monolingual children on overall phonological awareness tasks. The bilinguals performed significantly higher than monolinguals on all three types of phonological awareness tasks (segmentation, deletion, and blending). Second, there was a significant difference between the groups with respect to phonological units of the tasks. The bilinguals performed significantly better than monolinguals on the phonemic unit tasks, but two groups did not differ significantly on syllabic unit tasks. There was an interaction effect between unit size(syllables and phonemes) and group (bilinguals and monolinguals). Third, there were correlations for both bilingual and monolingual children between overall phonological awareness skills and word recognition skills.

• Phonetics and Speech Sciences
• /
• v.8 no.1
• /
• pp.9-17
• /
• 2016

The purpose of this work is to examine the voice onset times(VOTs) of the voiceless and voiced stops from the ten young male speakers of the Buckeye corpus[9]. The factors that are known to affect VOTs were also extracted, including the place of articulation, height of following vowels, location within word, presence of a preceding [s], status of the target word with respect to the content versus function word, presence of a syllabic stress, word frequency and speech rate. Findings from this work mostly agreed with those from earlier studies on English, but with some exceptions and new discoveries. We hope that this work can contribute to figuring out the nature and properties of the spontaneous speech of English.