A Study on the Aerodynamic Characteristics of Sustained Vowel and Continuous Speech Production of Female Voice Disorder Patients

표화영(Hwa Young Pyo) 한국언어청각임상학회 2019 Communication Sciences and Disorders Vol.24 No.2

배경 및 목적: 음성장애 화자를 평가할 때 연장모음과 연속발화가 보이는 특성이 서로 다름은 익히 알려져 있으나 공기역학적 측면에서의 이러한 연구는 매우 드물다. 이에 본 연구는 음성장애 화자의 공기역학적 특성이 연장모음과 연속발화 시 어떤 상관을 보이는지 알아보고자 하였다. 방법: 43명의 음성장애 여성화자로 하여금 /ㅏ/ 모음의 연장발성과 ‘가을’ 문단 낭독을 하도록 하였다. 이를 Phonatory Aerodynamic System의 Maximum Sustained Phonation과 Running Speech 프로그램을 이용하여 연장모음 및 연속발화의 공기역학적 측면을 분석하고 측정치 간의 상관성을 비교하였다. 결과: 공기역학적 측정치 중 시간 관련 영역 측정치는 연장모음과 연속발화간의 유의한 상관이 없었다. 기류량과 기류체적 관련 영역 측정치는 유의한 상관을 보이기는 했으나 그 상관의 정도가 높지는 않았다(r=-.521-.504). 논의 및 결론: 연구 결과, 공기역학적 측면에서 연속발화 측정치와 연장모음 측정치 간의 상관성은 유의하지 않거나 유의해도 상관성이 높지 않은 것으로 나타났다. 그러므로 임상적 의미가 있는 공기역학적 평가를 실시하기 위해서는 연장모음뿐만 아니라 연속발화 과제도 함께 고려해야 할 것이다. Objectives: Although it is well-known that the characteristics of sustained vowel and continuous speech are very different when evaluating voice problems, very little information has been reported on the aerodynamic aspects of these differences. The present study was performed to compare the aerodynamic characteristics of sustained vowel phonation and continuous speech production. Methods: Forty-three female patients with voice disorders produced a sustained vowel /a/ and continuous speech through paragraph reading, and their aerodynamic characteristics were analyzed using the ‘Maximum Sustained Phonation’ and ‘Running Speech’ programs of Phonatory Aerodynamic Systems (PAS). Results: There was no significant correlation between the sustained vowel and continuous speech in the time domain of the aerodynamic measurement results. However, there was a significant, but not high, correlation (r=-.521-.504) between the two speech tasks in the airflow rate and volume measurement results. Conclusion: We found that the correlation between the two tasks was significant, but not high. Therefore, in evaluating the aerodynamic characteristics of voice disorder patients, it is necessary to include continuous speech as well as a sustained vowel for meaningful and valid results.

한국 정상 화자의 캡스트럼과 스펙트럼 측정치의 성별과 발화 과제 효과

최성희,최철희 한국청각언어재활학회 2016 Audiology and Speech Research Vol.12 No.3

Purpose: The purpose of this study was to explore the effects of speech tasks (vowel vs. continuous speech) and gender on cepstral and spectral measurement. Methods: Fifty-one young adults with normal voice participated and four cepstral- and spectral parameters [cepstral peak prominence (CPP), CPPSD, L/H ratio, L/H ratio SD] were obtained using ADSVTM on seven Korean vowels (/a/, /ε/, /i/, /o/, / u/, / m/, /^/) and two standardized paragraphs (‘fall’ & ‘walk’). Results: Significant main effects of both gender and vowel type were found. Males had significantly greater mean CPPs than females and greater CPPs for low vowel /a/ than high vowels (/i/ and /u/). No effect of continuous speech type on CPPs and L/H ratio, but gender had significant effect on the L/H ratio (greater L/H ratio for males in both continuous speech tasks). In addition, vowels had significantly greater CPP and L/H ratio than continuous speech including various phonemes. There was no relationship between ‘walk’ and ‘fall’ in CPP whereas higher correlation was presented between ‘walk’ and ‘fall’ in L/H ratio. In normal voice, no evidence provided that /a/vowel had any relationships with both standardized paragraphs in CPP reflecting overall voice quality or severity of voice disorders. Conclusion: The results of this study indicate that cepstral- and spectral-based measures for normative data need to consider vowel type and gender. Furthermore, gender differences should also be considered for L/H ratio spectral-based measurement in continous speech.

발화 과제 유형에 따른 청지각적 평가와 기기적 평가의 상관성 연구

이옥분(Ok Bun Lee),권순복(Soon Bok Kwon) 한국언어치료학회 2012 言語治療硏究 Vol.21 No.2

The purpose of this study was to know the relationship between auditory-perceptual ratings and acoustic and aerodynamic evaluations according to sustained vowels and continuous speech. Fourty-one speakers' sustained vowel and reading samples with voice problems(palsy, polyp, MTD, nodules) were used for auditory perceptual judgement with 10-point scale. Five acoustic measures and two aerodynamic measures were extracted from sustained vowels. The results showed overall the significant correlation between continuous speech and sustained vowels(/a/, /i/, /u/) and GRBAS with r=.453~r=.758( p<0.01). The correlation between G, R, B in GRBAS and continuous speech was higher than sustained vowels(r=.758, r=.472, r=702, p<0.01). In the relation with acoustic evaluation, continuous speech had a weak correlation with jitter value only(r=.540, p<0.01). But three sustained vowels were mildly correlated with Fo, jitter, shimmer, NHR (r=-.432∼.697( p<0.01)) except for SFF. In correlation with aerodynamic evaluation, continuous speech and three sustained, both showed significant correlation from r=.352( p<0.05) to r=.627( p<0.01) except for perceptual judgement of sustained vowel for /u/ and subglottic pressure value.

음운 현상과 연속 발화에서의 단어 인지 : 종성중화 작용을 중심으로

김선미(Kim Sunmi),남기춘(Nam Kichun) 한국음성학회 2010 말소리와 음성과학 Vol.2 No.2

This study explores whether Koreans exploit their native coda-neutralization process when recognizing words in Korean continuous speech. According to the phonological rules in Korean, coda-neutralization process must come before the liaison process, as long as the latter(i.e. liaison process) occurs between ‘words’, which results in liaison-consonants being coda-neutralized ones such as /b/, /d/, or /g/, rather than non-neutralized ones like /p/, /t/, /k/, /?/, /?/, or /s/. Consequently, if Korean listeners use their native coda-neutralization rules when processing speech input, word recognition will be hampered when non-neutralized consonants precede vowel-initial targets. Word-spotting and word-monitoring tasks were conducted in Experiment 1 and 2, respectively. In both experiments, listeners recognized words faster and more accurately when vowel-initial target words were preceded by coda-neutralized consonants than when preceded by coda non-neutralized ones. The results show that Korean listeners exploit the coda-neutralization process when processing their native spoken language.

젊은 성인 및 노인 음성장애 환자의 연속발화시 공기역학적 특성 비교

표화영 한국콘텐츠학회 2019 한국콘텐츠학회논문지 Vol.19 No.12

This study was performed to compare the aerodynamic characteristics of young and elderly adult male patients with voice disorders during continuous speech. Aerodynamic measurements were obtained after 12 young male patients and 9 elderly male patients read a paragraph. The elderly group showed longer duration, lower airflow rate and air volume than the younger group, but the differences were not significant except phonation time. So, when interpreting the meaning of aerodynamic measures of elderly voice disorder patients in the aspects of airflow and air volume, it should take into account various conditions(e. g. reading materials, pulmonary functions) as well as age. 본 연구는 젊은 성인 음성장애 환자 집단과 노인 음성장애 환자 집단을 대상으로 문단낭독시 공기역학적 특성을 비교해보고자 시행되었다. 20-45세의 남성 음성장애 환자 12명과 60세 이상의 남성 음성장애 환자 9명이 연구에 참여하였다. PAS(Phonatory Aerodynamic System)의 Running Speech를 이용하여 문단을 낭독하게 하고 공기역학 측정치를 구하였다. 그 결과 노인집단은 젊은 성인집단보다 더 긴 지속시간과 더 적은 기류량 및 기류체적을 보였다. 그러나 그 차이는 발성지속시간에서만 유의하게 나타났다. 공기역학 측정치는 일반적으로 노인보다 젊은 성인이 모든 측면에서 비정상 소견을 보일 것이라고 예상하지만 본 연구결과 노인집단이 보인 측정치는 발성지속시간을 제외하고 기류량과 기류체적은 젊은 성인집단과 비슷한 수치를 보인 것으로 나타났다. 그러므로 노인 음성장애 환자의 연속발화시 공기역학 측정치가 갖는 의미를 해석할 때 기류와 기류체적은 연령뿐 아니라 발화자료, 폐 기능 등 다양한 요건을 함께 고려해야 한다.

한국어 연속음성인식 시스템 구현을 위한 형태소 단위의 발음 변화 모델링

정민화,이경님,Chung Minhwa,Lee Kyong-Nim 대한음성학회 2004 말소리 Vol.49 No.-

In this paper, we describe a cross-morpheme pronunciation variation model which is especially useful for constructing morpheme-based pronunciation lexicon to improve the performance of a Korean LVCSR. There are a lot of pronunciation variations occurring at morpheme boundaries in continuous speech. Since phonemic context together with morphological category and morpheme boundary information affect Korean pronunciation variations, we have distinguished phonological rules that can be applied to phonemes in within-morpheme and cross-morpheme. The results of 33K-morpheme Korean CSR experiments show that an absolute reduction of 1.45% in WER from the baseline performance of 18.42% WER was achieved by modeling proposed pronunciation variations with a possible multiple context-dependent pronunciation lexicon.

Robust Speech Detection Based on Useful Bands for Continuous Digit Speech over Telephone Networks

Ji, Mi-Kyongi,Suh, Young-Joo,Kim, Hoi-Rin,Kim, Sang-Hun The Acoustical Society of Korea 2003 韓國音響學會誌 Vol.22 No.e3

One of the most important problems in speech recognition is to detect the presence of speech in adverse environments. In other words, the accurate detection of speech boundary is critical to the performance of speech recognition. Furthermore the speech detection problem becomes severer when recognition systems are used over the telephone network, especially wireless network and noisy environment. Therefore this paper describes various speech detection algorithms for continuous digit recognition system used over wire/wireless telephone networks and we propose a algorithm in order to improve the robustness of speech detection using useful band selection under noisy telephone networks. In this paper, we compare some speech detection algorithms with the proposed one, and present experimental results done with various SNRs. The results show that the new algorithm outperforms the other speech detection methods.

말하기 관련 비교과 프로그램의 운영 현황과 특징

강연임(Gang Yeon im) 한국언어문학회 2017 한국언어문학 Vol.102 No.-

This thesis is studied about a management state and education result of extracurricular program, which is ‘Speech Clinic’ in Mokwon University. In the second chapter, it is examined a management state and detailed example of ‘Speech Clinic’. In third chapter, Operational performance and its expectation effectiveness, and a necessity of extracurricular program that is Curriculum connection show. In fourth chapter, it tries to find a future measures and its solution of extracurricular program ‘Speech Clinic’. However after standing out of speech, each university operates a speech lecture. But speech ability can not good better if one or two speech lecture could complete. And after take a class, there is a need of continuous education service. Therefore to fill the gap, it is used an extracurricular program that is connected with speech class. The extracurricular program is very useful to offer an intensive education service for students. ‘Speech Clinic’ in extracurricular program of Mokwon university is very useful that a subject professor counsels directly, so they know very well about subject’s character, student’s tendency, and contents of a class etc. And it can offer one-to-one education service, so students can receive solution of their speech problems directly and concretely. In ‘Speech Clinic’, it has a standardized manual of speech problems, and their professors have an workshop regularly. They share with various information about counselling, and try to solve the problem Collaboration. The problems of extracurricular program are its temporary management and short connection of class. So to gather an educational effect with a long-term perspective, it can not stop but manage with continuous support. And with a standardization work of counsel manual and professor’s steady studying, and cooperative discussion, it could be operated elastically to give a real help for students.

A Continuous Abnormal Speech Detection Method Based on Time Domain features Weighted

He Jun,Ji-chen Yang,Qing-hua Zhang,Guo-xi Sun,Jian-bing Xiong 보안공학연구지원센터 2014 International Journal of Control and Automation Vol.7 No.12

In this brief, a novel pathological continuous speech detection method based on time domain features weighted. First, different optimal threshold for time domain features, including zero crossing ratio, short-time energy and autocorrelation, are obtained from training speech data. Second, a difference evaluation technique is proposed, and with it, the difference of the same time domain feature selected from testing speech data and training speech data were obtained. Finally, to distinguish a given speech well, a novel weighting method based on difference evaluation for each kinds of time domain is employed, respectively. Experiments were conducted on the pathological speech database to prove the power and effectiveness of the proposed method. Results obtained shown that this method outperforms other early proposed time domain feature method, creating a more reliable technique for pathological continuous speech detection.

연속음성인식을 위한 Mel 주파수 켑스트럼 계수를 사용한 음성종속 화자식별

최재승(Jae-Seung Choi) 한국정보기술학회 2016 한국정보기술학회논문지 Vol.14 No.10

In order to perform a continuous speech recognition in the noisy environments, it is necessary to improve the noise performance improvement in conjugation with the speech recognition method. To build a continuous speech recognition model, this paper proposes a speaker identification algorithm which performs the continuous speech identification by using a Mel frequency cepstral coefficients and a multi-layer perceptron neural network. In the experiment of this paper, the proposed algorithm first improved the speaker identification rates in order to use the background noise suppression, then it was experimentally confirmed that the identification results by the proposed algorithm were improved effectively. From the experiment results, the maximum identification improvement was 5.80% for clean noise-free speech and 11.38% for white noise added speech compared with conventional method, respectively.