• The Journal of the Acoustical Society of Korea
• /
• v.17 no.5
• /
• pp.12-21
• /
• 1998

이 연구는 몇 가지 음성적 특징들이 한국어 연속 음성에서의 운율단위의 운율 단서 로 사용되어질 수 있음을 보여 준다. 구 단위의 운율 이론 체계에서 한국어의 운율단위를 악센트구와 억양구로 정의한 전선아(1993)의 연구 결과를 연속음성 자료에 도입하면서 운율 경계에서의 음성적 특징들을 살펴보았다. 연구 결과 악센트구와 억양구말에서는 피치 패턴 과 경계성조의 변화 뿐 아니라 단위말 음절의 길이 증가 현상이 두드러짐을 알 수 있었다. 또 악센트구와 억양구초에서는 모음으로 시작하는 음절의 경우, 모음 시작부에서의 성문음 화가 특징으로 나타난다. 운율 경계에서의 이런 음성적 특징들은 운율단위를 구분짓는 단서 로 이용될 수 있으며 또한 한국어 연속음성의 운율적 패턴을 이해할 수 있게 한다.

• Proceedings of the IEEK Conference
• /
• 2003.11b
• /
• pp.195-198
• /
• 2003

The accuracy of automatic speech recognition system depends on the presence of background noise and speaker variability such as sex, intonation of speech, and speaking rate. Specially, the speaking rate of both inter-speaker and intra-speaker is a serious cause of mis-recognition. In this paper, we propose the compensation method of the speaking rate by the ratio of each vowel's length in a phrase. First the number of feature vectors in a phrase is estimated by the information of speaking rate. Second, the estimated number of feature vectors is assigned to each syllable of the phrase according to the ratio of its vowel length. Finally, the process of feature vector extraction is operated by the number that assigned to each syllable in the phrase. As a result the accuracy of automatic speech recognition was improved using the proposed compensation method of the speaking rate.

• Speech Sciences
• /
• v.2
• /
• pp.135-147
• /
• 1997

This thesis describes pronunciation differences of vowel lengths between Koreans and Americans speaking English words and sentences. This study also analizes the reasons for these differences with the help of acoustic instruments. Sixteen sentences and eight words were selected as the experimental material. The informants for this study were 9 males; 3 Americans and 6 Koreans, who were asked to pronounce the test words and sentences five times. In this study, the acoustical analysis to measure duration was done through computer digital techniques. According to the results of the experiment, duration of 8 English vowels pronounced between Koreans and Americans shows very different features. When Koreans pronounce English vowels, the duration of the stressed vowel in the sentence-final position is much shorter than in other positions, such as in the sentence-initial and in word position. On the contrary, when Americans pronounce English vowels, the duration of the stressed vowel in the sentence-final position is much longer than in other positions. If the correlation between length and stress were to be studied in a more detailed manner, it would give fundamental help to the study of relation between stress and length.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.12a
• /
• pp.357-360
• /
• 2002

본 논문에서는 CCD 흑백 카메라를 이용하여 입력되는 인쇄체 한글 이미지의 문자를 인식하여 편집 가능한 텍스트 문서로 변환하는 시스템을 구현하였다. 문자 인식에 있어서 잡음에 강한 경계선 기울기 방법을 이용함으로써 문자의 구조적 특성에 근거한 윤곽선 정보를 추출할 수 있었다. 이를 이용하여 각 문자 이미지의 수평 및 수직 모음을 인지하고 6가지 유형으로 분류한 후, 자소 단위로 분리하고 최대 길이투영을 사용하여 모음을 인식하였다 분리된 자음은 경계선이 변화되는 위상의 형태를 미리 저장된 표준패턴과 비교하여 인식하였다. 인식된 문자는 KS 한글 완성형 코드로 문서 편집기에 출력되어 사용자에 제공되는 시스템을 구현하였다.

• Proceedings of the KSPS conference
• /
• 2005.11a
• /
• pp.91-95
• /
• 2005

The aim of this paper is to investigate the effects of postvocalic voicing(Contrasting voiceless fricative and affricate with voiced fricative and affricate) on vowel duration. In particular we focused on the durational differences between vowels followed by voiceless and voiced consonants across three groups of speakers: English speakers, English bilinguals and Korean learners of English. the result of experimental I showed that durations of vowels preceding voiced fricative and affricates as well as voiced stops are significantly longer than those preceding voiceless counterparts. Experiment Ⅱ indicated that as the subjects exposed themselves longer to English speaking society, their pronunciation was increasingly similar to those of English native speakers.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.27 no.4
• /
• pp.517-523
• /
• 2000

Nasopharyngeal closure is a sphincter mechanism between the activities of the soft palate, lateral pharyngeal wall and the posterior pharyngeal wall, which divides the oral cavity and the nasal cavity. It participates in physiological activities such as swallowing, breathing and pronunciation. In case of an error in this mechanism, it is called a nasopharyngeal incompetence. The causes of this error are defects in (1) length, function, posture of the soft palate (2) depth and width of the nasopharynx, (3) activity of the posterior and lateral pharyngeal wall. The purpose of this study is to analyze the nasopharynx of cleft palate patients using lateral cephalograms and at the same time, evaluate the degree of hypernasality of each vowels to find its relationship with nasopharyngeal incompetence. The following results were obtained: 1. The length of the soft palate was markedly short than normal. 2. The adequate ratio was smaller than the normal value. 3. As the adequate ratio decreased, when articulating vowels, anatomic mVPI increased. 4. When articulating each vowels, anatomic VPI was in proportion with the degree of hypernasality. 5. The degree of hypernasality was greater in high vowels(/i/, /u/) than low vowel(/a/). From the above results, it can be concluded that in cleft palate patients, lateral cephalograms can be used effectively in diagnosing and evaluating nasopharyngeal incompetence. The anatomic structure of the nasopharynx has close relation to the degree of hypernasality.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06e
• /
• pp.89-92
• /
• 1998

영어의 선행모음의 길이 특성이 어말 자음의 유.무성 인지에 어떤 영향을 미치는지에 관한 선행 연구의 후행연구로서 후행하는 폐쇄음의 자음적 특성은 어떠한 실마리로 적용하는지 한국어 화자들을 대상으로 한 인지실험을 통해 알아보았다. 실험결과, 한국어 모국어 화자들에게 어말 폐쇄음의 자음적 특성은 자음의 유.무성 인지에 중요한 실마리로 작용하지 못하고 있다는 사실을 알 수 있었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.08a
• /
• pp.188-191
• /
• 1998

호주 영어의 단어 중 한국인과 일본인이 구별하기 어려운 'work', 'walk', 'wok' 세 단어에 대해 다룬다. 우선 한국인과 일본인이 이들 세 단어를 어떻게 조음할 것인가를 예측하고, 실제로 한국인과 일본인 화자들이 조음한 자료를 가지고 모듬의 음질과 길이 두가지 측면으로 나누어 예측한 사항과 비교하고 또 이들 단어를 조음할 때 한국인과 일본인 사이에 어떤 음성 음운적 특징이 나타나는지도 비교 분석한다. 결과는 모국어의 음성, 음운적 영향이 목표어의 음소를 습득하고 발음하는데에 영향을 미친다는 사실을 보여주고 있다. 하지만 wok/wo k/의 예는 언어 보편적인 요소가 L2의 음소를 습득하는데 있어 또 하나의 요소로 작용하고 있음도 알려주고 있다.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.2
• /
• pp.88-95
• /
• 2006

In this Paper. a method for improving the performance of automatic speech recognition (ASR) system for conversational speech is proposed. which mainly focuses on increasing the robustness against the rapidly speaking utterances. The proposed method doesn't require an additional speech recognition task to represent speaking rate quantitatively. Energy distribution for special bands is employed to detect the vowel regions, the number of vowels Per unit second is then computed as speaking rate. To improve the Performance for fast speech. in the pervious methods. a sequence of the feature vectors is expanded by a given scaling factor, which is computed by a ratio between the standard phoneme duration and the measured one. However, in the method proposed herein. utterances are classified by their speaking rates. and the scaling factor is determined individually for each class. In this procedure, a maximum likelihood criterion is employed. By the results from the ASR experiments devised for the 10-digits mobile phone number. it is confirmed that the overall error rate was reduced by $17.8\%$ when the proposed method is employed

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.16 no.2
• /
• pp.180-195
• /
• 1991

In this paper, we studied on new method of recognition using structural information to recognize character pattern in orginal shape of Hangeul. First, for the purpose of knowing location of character in input image. it processed Making block. Second, after we investigated. whether vertical vowel exited or not in character image accordingly the center of gravity of Hangeul. each character was classified into Type of Hangeul by searching location and length for horizontal vowel and short pole. Last, we processed it by means of template matching which calculate Uclid's distance on each Jaso in accordance to type classified. This paper made an experiment on 2350 characters and obtained 98.3% classifing rate and 95.2% recognizing rate.