한국어 경북방언 의문문의 통사구조와 운율구조의 관계에 대하여

윤원희,박선우,엄홍준,김기태,김용하 현대문법학회 2019 현대문법연구 Vol.101 No.-

Morphosyntactic and prosodic information is accessed by native speakers of North Gyeongsang Korean when interrogatives are interpreted. The present study investigates the interface between these structures. To do so, the study analyzes the syntactic and prosodic structures of the dialect’s yes/no and wh-questions, and then examines the rate of comprehension and acceptance of the two types of the interrogatives in a perception test. The prosodic structures in the test are modified by transplantation, the results of which allow us to find out the following. First, presented with the interrogatives whose syntactic and prosodic structures did not match, the native subjects of Gyeongsang Korean relied more on the prosodic structures than on the syntactic ones. Second, changes in prosodic structures had a strong influence on simple sentences, but relatively less so on complex sentences. These results lead to the conclusion that prosodic structures are the decisive factor in syntactic interpretations, and, accordingly, are intricately intertwined with the syntactic structures during the processing of interrogatives.

폐쇄음 음향 단서의 다차원 표현과 상관관계 분석

윤원희,Yun, Weon-Hee 대한음성학회 2005 말소리 Vol.55 No.-

The purpose of this paper is to represent values of acoustic cues for Korean oral stops in the multi-dimensional space, and to attempt to find possible relationships among acoustic cues through correlation analyses. The acoustic cues used for differentiation of 3 types of Korean stops are closure duration, voice onset time and fundamental frequency of a vowel after a stop. The values of these cues are plotted in the two and three dimensional space to see what the critical cues are for separation of different types of stops. Correlation coefficient analyses show that multi-variate approach to statistical analysis is legitimate, and that there are statistically significant relationships among acoustic cues but Oey are not strong enough to make the conjecture that there is a possible relationship among the articulatory or laryngeal mechanisms employed by the acoustic cues.

경상 방언 내포문 의문사의 작용역 범주 지각 양상과 반응 속도 연구

윤원희 한국음성학회 2023 말소리와 음성과학 Vol.15 No.2

This study investigated the response time and patterns of categorical perception of the wh-scope of an embedded clause with the non-bridge verb, “gung-geum hada ‘wonder’,” in the matrix verb phrase in Gyeongsang Korean. Using the same procedure as Yun (2022), 72 responses and response times for each stimulus were collected from 24 participants over the course of three trials. The stimuli were recorded readings of 40 speakers (20 male, 20 female). Context was provided to induce a matrix scope interpretation of the embedded wh-phrase in the target sentence. We sorted the 40 stimuli according to the number of matrix scope responses each received, and charted the response times for each stimulus. Although there was considerable overlap for the different types of wh-scope interpretations, there was a clear difference in categorical perception between the matrix and embedded scopes. The 24 participants also differed in their categorical perceptions. The results suggested that response time and wh-scope interpretation were not directly related and that two main weighted factors affected wh-scope interpretation: morpho-syntactic constraints and prosodic structural integrity. The weighting of each of these factors was inversely correlated and varied among subjects.

Automatic Segmentation of Korean Oral Stops for Automatic Speech Recognition

윤원희 한국외국어대학교 언어연구소 2004 언어와 언어학 Vol.0 No.34

한국어 폐쇄음의 음성인식에 사용될 폐쇄 구간과 Voice Onset Time (VOT) 구간을 얻어내기 위하여 자동으로 폐쇄음의 폐쇄구간과 폐쇄음 뒤 모음 시작점을 분할하는 방법을 제안한다. 한국어 폐쇄음은 같은 조음위치에서 세 가지 Phonation type을 갖는다. 경음은 연음이나 격음보다 긴 폐쇄 구간을 가지고 있고, 격음은 경음이나 연음보다 긴 VOT를 가지고 있다는 음성학적 연구 성과에 비추어, 먼저 연속음성 데이터베이스인 KAIST 음성데이터에서 628개의 문장을 hand labelling 한 후, 통계 실험을 하였다. 이 실험은 문맥의 다양성에도 불구하고 폐쇄구간과 VOT의 특성이 phonation type에 따라 기존 음성학 연구에서의 결과가 같은지를 알아보기 위함이다. 실험 결과, 문맥의 다양성에도 불구하고 기존 연구 결과와 일치 함을 보였다, 이러한 결과는 음성인식에서 폐쇄음을 구분하기 위하여 폐쇄구간과 VOT의 두 가지 음향단서를 사용할 수 있음을 입증하는 것이다. 음성인식기에서 폐쇄음의 phonation type을 구별하기 위한 폐쇄구간과 VOT를 자동으로 추출할 수 있는 자동 분할 방법을 개발하였다. 클러스터링과 두 연속된 창에서의 로그 파워의 차이값의 local peak를 계산함에 의해서 이러한 구간들의 분할을 할 수 있다.

인성교육 프로그램이 대학생활만족과 이타주의행동에 미치는 영향

윤원희,오삼규,조영빈,윤상기 한국경영교육학회 2017 한국경영교육학회 학술발표대회논문집 Vol.2017 No.11

영어 모음 포먼트 군집화 분석

윤원희 한국영미어문학회 2020 영미어문학 Vol.- No.137

This study aims to investigate how vowels are categorized into a given number of clusters using an unsupervised machine learning technique called the k-means clustering algorithm. The results of matching clusters to vowel types can explain to what extent the categorizations of vowels are based purely on parameters obtained from the data. The parameters used for the clustering are the first and the second formants in bark. The dataset used in this study is from Hillenbrand et al. (1995). The target vowels were restricted to those produced by adult males. The k was increased stepwise from 3 to 9 to see what phonological features appeared to be active in separating a new cluster from the originating cluster. It was found that the feature [±back] is one of the primary features that differentiates vowel classes and the feature [±high] is used in subsequent separations of the vowel classes. It was also found that high confusability in the perception of vowels is attributable to overlapping distributions of the samples. Vowel duration is critical in clarifying /ɔ/ and /ʌ/, and /ɛ/ and /æ/ as was reported in Hillendbrand et al. (2000). In our study, /æ/ was the second last and /ɔ/ was the last cluster separated from /ɛ, æ/ and /ʌ, ɔ/ clusters, respectively.

Reanalysis of Vowel Formants in the Buckeye Corpus

윤원희 언어과학회 2020 언어과학연구 Vol.0 No.93

Vowel formants are an important piece of information for mapping acoustic signals to a specific position in a vowel space defined by cardinal vowels. Each formant of a vowel can be obtained by averaging multiple formant values calculated from sliding analysis windows applied over the signals in the vowel area or simply by taking the formant value at a time point. Depending on which part of the vowel span is used, the formant values can change and as a result its position in the vowel space can also change. Vowels in a large corpus of spontaneous speech can be affected by various uncontrolled factors, which brings about large deviations in formants. In particular, when averaging multiple values from a token, the internal variation in each vowel token is easily ignored. This paper shows that internal variations can affect the final results of the vowel formants that represent a specific vowel type. This is particularly the case with the first formant of non-low vowels which tends to be more affected than with low vowels.

조선의 영웅에 대한 기억의 형성과 변용 -김덕령, 임경업 두 장군을 중심으로

윤원희 한국역사민속학회 2023 역사민속학 Vol.- No.64

Kim Deok-ryeong and Lim Gyeong-up, heroes who were heroes but died tragically, were compared to the Chinese Song Dynasty general Akbi. They performed well, but their final appearances are similar. In other words, they are a general, but they did not die on the battlefield but was falsely framed by a traitor. However, if you look closely at the records of the deaths of Kim Deok-ryeong and Lim Kyung-up, it is different from the stories told in later generations. The most different part of them was that the kings of the time actively drove them to death rather than regretting their deaths. It began with “below” which actively formed the memories of the heroes' unjust deaths. They could not remember openly because Kim Deok-ryeong and Lim Kyung-up died in a treason case, but they inherited their memories among themselves. In particular, they tried to fulfill their wishes for life by emphasizing the Han elements and releasing their resentments. After a while, extreme disasters occur, and it was “exoneration” that the kings devised as one of the ways to solve these disasters. In addition, it was bureaucrats who grew up hearing memories from below that delivered Kim Deok-ryeong and Lim Gyeong-up to the king as objects of “exoneration.” Although exoneration was not easy to do, these heroes were recovered after discussions between the king and officials. However, it did not stop there but transformed the memory so that later kings could remember these heroes as loyal subjects for their needs. In other words, the heroes who died reversely were now remembered as loyalists. These changes in memory contain the expectations of each subject surrounding memory and the needs of the time. 영웅이었지만 역적으로 몰려 죽은 영웅들인 김덕령과 임경업은 중국 송대 명장 악비에 비유되었다. 이들은 활약상이 뛰어나기도 했지만, 최후의 모습들이 유사하다. 즉, 명장이었지만 전쟁터에서죽지 않고 간신에 의해 억울하게 모함당해 죽었다는 점이다. 그런데 김덕령과 임경업의 죽음에 대한 기록을 자세히 살펴보면 후대에 전해진 이야기들과 다르다. 그 중 가장 다르게 기억된 부분은당대 왕들이 영웅들의 죽음에 대해 애석해 하는 모습으로 기록되었는데, 실상은 적극적으로 영웅들을 죽음으로 몰아간 점도 보인다는 것이다. 영웅들의 억울한 죽음에 대한 기억을 적극적으로 형성한 것은 아래, 즉 민간에서 시작되었다. 이들은 김덕령과 임경업이 역모 사건에 얽혀 죽었기 때문에 드러내놓고 기억할 수는 없었지만, 자신들끼리 기억을 계승했다. 특히 한적인 요소를 강조하며 한을 풀어줌으로써 자신들의 삶의 소원을이루고자 했다. 시간이 지난 뒤 극심한 재난이 발생하고, 이 재난들을 해결할 수 있는 방법 중 하나로 왕들이 고안한 것이 바로 伸冤이었다. 그리고 김덕령과 임경업을 신원의 대상으로 왕에게 전달한 것이 민간의 기억을 듣고 자란 사족들이었다. 신원 자체가 쉽게 할 수 없는 일이었지만 왕과관료들의 논의 끝에 이 영웅들은 신원되었다. 그런데 여기서 그치지 않고 후대 왕들이 자신들의 필요를 위해 이 영웅들을 충신으로 기억하게끔 기억을 변용시켰다. 즉, 역적으로 죽은 영웅들이 이제충신으로 기억되었다. 이와 같은 기억의 변화들은 기억을 둘러싼 각 주체들의 기대와 당대의 필요가 담겨 있다.

경상 방언 의문문 작용역의 지각 구분

윤원희 한국음성학회 2022 말소리와 음성과학 Vol.14 No.2

A wh-phrase positioned in an embedded clause can be interpreted as having a matrix scope if the sentence is produced with proper prosodic structures such as the wh-intonation. In a previous experiment, a sentence with a wh-phrase in an embedded clause was given to 40 speakers of Gyeongsang Korean. A script containing the sentence was provided to induce a matrix scope interpretation for the wh-phrase. These 40 utterances were prepared as stimuli for a perception test to verify whether the wh-phrases in the stimuli were perceived as having matrix scopes. Each utterance was played thrice to 24 subjects. The results showed that more than half of the 72 responses indicated a preference for an embedded scope rather than a matrix scope in 20 of the utterances. A multiple linear regression analysis showed that the matrix scope responses were best predicted by the magnitude of the pitch prominence in a prosodic word consisting of an embedded verb and a complementizer. The pitch prominence was calculated by subtracting the fundamental frequency (F0) at the right edge of the prosodic word from the peak F0 in the same prosodic word. The smaller the magnitude, the more matrix responses there were. These results suggest that the categorical perception of wh-scopes is based on the magnitude of pitch prominence.

Pronunciation Dictionary of Korean learners of English for Automatic Pronunciation Scoring: A Theoretical Approach

윤원희 한국외국어대학교 언어연구소 2009 언어와 언어학 Vol.0 No.45

We review the methods for machine scoring measuring pronunciation accuracy of non-native speech and suggest a theoretical basis for a possible modification of a dictionary using a multiple pronunciation variation for Korean learners of English. Machine scoring systems use a dictionary including variations originated from the systematic errors. However, the variations are mostly word internal errors. We reveal theoretically that errors across word boundaries should be included in the dictionary. Since these errors are usually conditioned by the first segment of the following word, the dictionary must be constrained not to generate unnecessary search network and to prevent erroneous phone recognition.