FIR filtering에 의한 끝점추출에 관한 연구

이창영 한국음성과학회 1999 음성과학 Vol.5 No.1

종전의 끝점추출은 주어진 입력 음성신호 자체의 에너지가 어느 문턱값을 넘는가 하는 판별에 의해 이루어지는데, 저주파의 배경잡음이 있는 경우 그 방법은 실패할 가능성이 높다. 이를 극복하기 위해 문턱값을 증가시키면 모음/으/와 같이 에너지가 작은 음성들이 끝점추출에서 탈락하는 일이 발생한다. 이처럼, 종전의 방법에서 단지 문턱값의 조절만으로는 끝점추출에 한계가 있는 것이다. 본 연구에서는 1차 FIR filtering으로 음성신호를 변환한 후, 이 변화된 신호에 대해 종전과 같은 끝점추출을 수행하는 방법을 제시하였고, 그 결과 기존의 방법에 비해 10%미만의 에러로 끝점추출이 수행됨을 확인하였다. 이러한 변환에 의해 음성 신호에서 (진폭 X 주파수)가 동시에 큰 성분만이 걸러진다. 즉, 에너지가 작은 잡음, 또는 에너지는 어느 정도 커도 주파수가 작은 배경잡음이 효과적으로 제거되는 것이다. ETRI의 445 단어 음성 DB에 본 연구에서 고안된 방법을 적용한 결고, 저주파 배경잡음이 변환에 의해 평평한 noise level로 바뀌어, 종전의 방법에 의해 구분이 어려운 경우에서도 끝점추출이 올바로 수행됨을 확인하였다. This paper provides a method for speech detection. After first order FIR filtering on the speech signals, we applied the conventional method of endpoint detection which utilizes the energy as the criterion in separating signals from background noise. By FIR filtering, only the Fourier components with large values of [amplitude x frequency] become significant in energy profile. By applying this procedure to the 445-words database constructed fROM ETRI, we confirmed that the low-amplitude noise and/or the low-frequency noise are separated clearly from the speech signals, thereby enhancing the feasibility of ideal endpoint detections. Keywords: speech detection, endpoint detection, speech recognition, analysis

한국인과 원어민 영어 발화의 F0 고저 범위 차이와 발음 평가에 있어서 그 역할

이석재,조철현,문선영 한국음성과학회 2003 음성과학 Vol.10 No.4

This study examines the high-low pitch range differences between Korean and native English speakers' utterances of 36 English sentences, and explores the role of the pitch fluctuation range in the pronunciation assessment. Findings are: i) Korean speakers implement half as much pitch fluctuation as the native speakers with the achievement rate of, on an average, 47%. Especially, pitch fluctuation range is below the average in the sentence types of exclamation, yes-no & wh-questions, ii) the reason why Korean speakers' pitch fluctuation range is only half the range of the native speakers is that Korean speakers do not raise the pitch at certain sentence positions so much as the native speakers, and iii) the range of pitch fluctuation plays an important role in distinguishing the utterances of the native speakers from those of Korean speakers, but does not in rating the English utterances among Korean speakers.

한국인의 외국어 발화오류검출 음성인식기에서 청취판단과 상관관계가 높은 기계 스코어링 기법

배민영,권철홍 한국음성과학회 2004 음성과학 Vol.11 No.2

An automatic pronunciation correction system provides users with correction guidelines for each pronunciation error. For this purpose, we develop a speech recognition system which automatically classifies pronunciation errors when Koreans speak a foreign language. In this paper, we propose a machine scoring method for automatic assessment of pronunciation quality by the speech recognizer. Scores obtained from an expert human listener are used as the reference to evaluate the different machine scores and to provide targets when training some of algorithms. We use a log-likelihood score and a normalized log-likelihood score as machine scoring methods. Experimental results show that the normalized log-likelihood score had higher correlation with human scores than that obtained using the log-likelihood score.

설단 폐쇄음의 목표 F2 값 : 한국어, 영어, 불어의 비교

오은진 한국음성과학회 2003 음성과학 Vol.10 No.4

The aim of this study was to estimate the target F2 values of the coronal plain stop in Korean and the degree of deviation from the target in the context of various vowels, and to compare the results of Korean regarding the coronal stop with those of English and French. An acoustic analysis showed that the mean F2 value of the Korean coronal stop produced by 10 male speakers was 1,855 Hz and the deviation from the target was 94 Hz in the context of [j], 204 Hz in the context of [u], and 407 Hz in the context of [o], The target F2s of the coronal stop were the highest in English (1,929 Hz) and the lowest in French (1,662 Hz), and the deviation from the targets in the context of the high back vowel was the largest in French (257 Hz) and the smallest in English (73 Hz).

한국 성인의 정상 음성에 관한 기본 음성 측정치 연구

표화영,심현섭,송윤경,윤영선,이은경,임성은,하현령,최홍식 한국음성과학회 2002 음성과학 Vol.9 No.2

Our present study was performed to investigate acoustically the Korean normal adults'; voices, with enough large number of subjects to be reliable. 120 Korean normal adults (60 males and 60 females) of the age of 20 to 39 years produced sustained three vowels, /a/, /i/, and /u/ and read a part of 'Taking a Walk' paragraph, and by analyzing them acoustically with MDVP of CSL, we could get the fundamental frequency (F_0), jitter, shimmer and NHR of sustained vowels; speaking fundamental frequency (SF_0), highest speaking frequency (SFhi), lowest speaking frequency (SFIo) of continuous speech. As results, on the average, male voices showed 118.1~122.6 ㎐ in F_0, 0.467~0.659% in jitter, 1.538~2.674% in shimmer, 0.117~0.133 in NHR, 120.8 ㎐ in SF_0, 183.2 ㎐ in SFhi, 82.6 ㎐ in SFIo. And, female voices showed 211.6~220.3 ㎐ in F0, 0.678~0.935% in jitter, 1.478~2.582% in shimmer, 0.098~0.114 in NHR, 217.1 ㎐ in SF_0, 340.9 ㎐ in SFhi, 136.0 ㎐ in SFIo. Among the 7 parameters, every parameters except shimmer showed the significant difference between male and female voices. And, when we compared the three vowels, they showed significant differences one another in shimmer and NHR of both genders, but not in F_0 of males and jitter of females.

청력손실아동의 재활에 관한 고찰

김진숙,이정학 한국음성과학회 1998 음성과학 Vol.4 No.2

지금까지 청각기관의 발달과정과 진단법에 따른 유소아의 청각재활에 관해 알아보았다. 태아기 20주에 기능이 가능한 청각기관은 출생 시 이미 언어를 습득할 준비가 되어있어 다양한 청각자극을 통해 언어를 발달시킬 수 있다. 그러므로 청각 손실은 곧 언어장애와 그에 따른 발달 및 사회성장애와 연결되어 청각손실의 조기 발견과 조기재활은 매우 중요하다. 조기발견을 하기 위하여 개발된 유소아의 객관적 혹은 주관적 청각검사의 활용이 객관적 검사에 절대적으로 의존하고 있는 한국적 상황에서는 시급한 실정이다. 따라서 본고에서 제시된 여러 가지 주관적 청력평가의 방법론 등이 널리 보급되고 표준화 된 한국식 평가법에 대한 활발한 연구가 진행되기를 바라는 바이다. 더구나 기계적인 청력보강으로 청각자극을 극대화시키고 훈련을 통해 언어능력을 증강시키는 조기재활은 현실적으로 참고할 자료조차 마땅치 않다. 에 본고는 유소아의 광범위한 청각재활의 정의와 목적 등을 세우고, 더불어 그 훈련방법의 기준 및 방법적 체재를 마련하고자 하였다. 물론 모든 청력손실의 정도와 유형을 모두 포함하기에는 다소 무기가 있지만, 청각재활의 고유영역인 청능훈련의 대표적 두 가지 기법, 즉 대화기법과 듣기훈련법을 소개하고 그 중 듣기훈련법은 다섯 단계별 특성을 위주로 유소아 청각재활에 전체적 윤곽을 그려보았다. 소리에 대한 청각감지, 구별, 확인, 인지, 이해 등의 다섯 단계로 구분되어 있는 훈련법은 아동의 수행능력에 따라 탄력적으로 실시되어야 할 것 이다. 또한 자녀의 청력손실을 발견한 부모의 심리 및 반응을 분석하여 청각재활에 가장 큰 정신적 지주가 되는 상담교육의 효율성을 증가시키고자 하였다. 덧붙여 보다 나은 청각재활의 방법 및 기준확립을 위하여 많은 연구가 후속되어야 할 것이다. The human baby appears to be born with preexistent knowledge of language. This specialized neural structure in the brain awaits auditory experience with language to trigger it into functioning. The auditory-linked acquisition of language is a time-locked function related to early maturational periods in the infant's life. The longer auditory language stimulation is delayed, the less efficient the language facility will be. It is for these reasons that it is urgent to fight the hearing problems of children with all the skill, knowledge and insights of which we are capable, the so called "rehabilitative process". An understanding of the timetable and the origin of prenatal to early life development of auditory mechanism will help in planning the aural rehabilitation. Further interests and studies are needed to establish the systematic structure of rehabilitative audiology. Keywords: auditory experience, rehabilitative audiology, hearing impaired

음장과 외이도 내부에서의 음성 비교

허승덕,김리석,고도흥,이정학 한국음성과학회 2000 음성과학 Vol.7 No.4

외이도 내부의 음성은 음원에서부터 매질인 음장 그리고 청각 기관의 말단인 외이도까지 다양한 조건에 의해 변화하였는데, 발성기관과 음장의 상호 작용에 의해 모든 대상 모음의 F_1, F_2와 조음위치가 높은 모음의 F_3, F_4, F_5가 음장의 감쇄에 의해서는 /a/의 F_4, F_5와 /i/와 /u/의 F_3, F_4, F_5가 머리와 이개의 간섭의 의해 /a/의 F_1, F_2, /i/의 F_2가 그리고 외이도의 구조 및 공명에 의해 /a/의 F_1, F_3와 /i/의 F_3, F_5가 각각 낮아지거나 높아졌으며, 조음 위치가 높은 모음에서는 전설보다 후설모음에서 그리고 조음위치가 높은 모음보다 낮은 모음에서 더 큰 변화를 보였다. 외이도 내부에서의 음성은 음장을 지나 온 음원에 대한 외이도의 좁아진 공간뿐만 아니라 구강 공간의 크기, 구강 전반부 및 인가의좁힘 등의 발화 요인과 음장 조건 등에 의해서도 영향을 받고 있었으며, 외이도 공명 주파수를 중심으로 각각의 모음 앞쪽과 뒤쪽 포먼트 주파수 및 강도 변화 양상으로 보아 음성의 인지보다는 음성의 식별과 관련된 것으로 추정된다. 더욱이 본 연구결과의 활용을 위해 보다 많은 실험 단어와 음성 분석 지료를 사용한 연구가 필요하겠다. The purpose of this study was to examine some acoustic chracteristics in the ear canal. It was assumed that a sound outside the external auditory canal could be different from the sound inside the external auditory canal. The acoustic signals were captured by a probe microphone placed at a distance within 1 cm from the tympanic membrane, and a reference microphone was placed over the upper pinna. Three vowels /a/, /i/, /u/ were recorded from a normal adult male speaker. The parameters such as the formant frequency (F1~F5) and the peak intensity were measured using a speech analyser, PCquirer. It was found that the entering part of the external auditory canal functions as a narrowing point as to the speech that passes through the free field. Results show that acoustic characteristics were chaned for speech discrimination rather than speech perception. Keywords : free field, external auditory canal, speech discrimination

한국어/영어 이중언어사용 아동의 한국어 문장이해 : 조사, 의미, 어순 단서의 활용을 중심으로

황민아 한국음성과학회 2003 음성과학 Vol.10 No.4

The purpose of the present study was to investigate the sentence comprehension strategies used by Korean/English bilingual children when they listened to sentences of their first language, i.e., Korean. The framework of competition model was employed to analyze the influence of the second language, i.e., English, during comprehension of Korean sentences. The participants included 10 bilingual children (ages 7;4-13;0) and 20 Korean-speaking monolingual children(ages 5;7-6;10) with similar levels of development in Korean language as bilingual children. In an act-out procedure, the children were asked to determine the agent in sentences composed of two nouns and a verb with varying conditions of three cues(case-marker, animacy, and word-order). The results revealed that both groups of children used the case marker cues as the strongest cue among the three. The bilingual children relied on case-marker cues even more than the monolingual children. However, the bilingual children used animacy cues significantly less than the monolingual children. There were no significant differences between the groups in the use of word-order cues. The bilingual children appeared less effective in utilizing animacy cues in Korean sentence comprehension due to the backward transfer from English where the cue strength of animacy is very weak. The influence of the second language on the development of the first language in bilingual children was discussed.

한국어 특수조사 '_만'에 연계된 초점의 음향음성학적 특성

최재웅,전윤실,장윤,박순복,김기호 한국음성과학회 1999 음성과학 Vol.5 No.2

The purpose of this paper is to investigate the phonetic characteristics of 'focus' phrases accociated with the particle '-man' in Korean. The particle '-man' is a bound morpheme which, like other postpositions such as the subject marker'-ka' and the object marker '1??1', the so-called 'case markers' in Korean, typically attaches to a nuun (phrase). The semantics of '-man' roughly corresponds to that of only, its counterpart in English, and is thus classified as a 'delimiter' (Yang 1973). It is assumed in this paper that '-man-, like only in English, shoud have a 'focus' associated with it (von Stechow 1991, Rooth 1992). In general, '-man' attached pharases get the focus, but sometimes the association is not clean-cut, especially in the cases of emphatic use of '-man' or when the context strongly favors other phrase as the focus (Choe 1996). In this paper, we compare the phonetic characteristics of the '-man' marked phrases with those to which '-ka'/'-1??1' is attached, and conclude that the focused '-man' phrases show higher fundamental frequencies than their equally focused 'case'-~marked counterparts. However, when the context clearly forces the focus to fall on phrases other than the '-man' or '-ka'/'-1??1' attached ones, there is no meanlingful difference in fundamental frequency between the '-man' and '-ka'/'-1??1' attached phrases. We also compare the phonetic characteristics of the regular use of '-man' with those of the emphatic '-man'. According to our experiments, the emphatic '-man' does not bring forth its phonetic effects, namely, higher fundamental frequencies, on the '-man' attached words or phrases but rather in various other ways such as higher fundamental frequencies in '-man', lengthening of the following word-initial syllable, or the inclusion of the following word in the same accentual phrase. Finally, it is claimed that '-man' associated focus phenomena, especially the emphatic use of '-man', show some typical acoustic characteristics of the other well-known focus phenomena, namely, wh-interrogatives. Key words:korean particle '-man', foucs, Fo, accentual phrase

다중음성치료기법(SK-MVTT)이 성대결절 환자의 음성개선에 미치는 효과 : 액센트 기법과의 비교 연구 Comparison with Accent Method

김성태,정옥란 한국음성과학회 2004 음성과학 Vol.11 No.4

Vocal nodule is one of the representative chronic diseases of vocal folds, and it can be cured by voice therapy. However, the existing therapeutic methods about vocal nodule are in great demand. The purpose of this study was to evaluate the effect of therapeutic methods between Accent method and SK-MVTT (Seong-Tae Kim's multiple voice therapy technique), which was designed by the author. We identified 40 females, who diagnosed having vocal nodules, aged from 21 to 52 years (mean age: 40 years). Twenty females were treated by the SK-MVTT and the other 20 females the Accent method. All subjects received 12 sessions of treatment, and were evaluated after finishing the 6th and the 12th session. The results showed that the SK-MVTT produced a better results compared to the Accent method. The SK-MVTT was better especially at the initial stage of voice therapy compared to the Accent method. In this study, we can suggest that SK-MVTT may be useful in improving the voice qualities of vocal nodule patients. However, more data should be collected and evaluated before it can widely be used in other clinics.