• Journal of Digital Contents Society
• /
• v.18 no.6
• /
• pp.1049-1056
• /
• 2017

This study focused on the changes in the voices in determining the effects of speech training. This study aimed to make more visible and scientific evaluation of the changes in the voices among the substantial effects obtained from speech training. As a result, some objective differences from before the speech training could be found in the voice of every learner. Each learner showed gradual technical improvement in a variety of vocal elements, including resonance and timbre, accuracy of pronunciation, pause; that is, the voice became more powerful, more accurate pronounced, more pausing and more stable than before the speech training. This study determined if speech training could change a voice and the results are expected to help speech learners participate actively in speech training and see their speech ability improved.

• Journal of Korea Society of Industrial Information Systems
• /
• v.25 no.4
• /
• pp.9-16
• /
• 2020

Pronunciation recognition systems that use only video information and ignore voice information can be applied to various customized services. In this paper, we develop a system that applies a Bayesian classifier to distinguish Korean vowels via lip shapes in images. We extract feature vectors from the lip shapes of facial images and apply them to the designed machine learning model. Our experiments show that the system's recognition rate is 94% for the pronunciation of 'A', and the system's average recognition rate is approximately 84%, which is higher than that of the CNN tested for comparison. Our results show that our Bayesian classification method with feature values from lip region landmarks is efficient on a small training set. Therefore, it can be used for application development on limited hardware such as mobile devices.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.19 no.2
• /
• pp.202-209
• /
• 2018

This study analyzed the effects of the nutrient intake on oral health and chewing difficulty according to the age group. The subjects were 5,855 participants of the third Korea National Health and Nutrition Examination Survey(KNHANES VI), 2015, Korea Centers for Disease Control and prevention and aged 20 years and over. The data were analyzed using SPSS Ver 21.0, classified as the difficulty in chewing group (DC) and no difficulty in chewing group (NDC). As a result, the DC rates were 5 times higher in the 60+ year age group (39.5%) than in the 20-39 year age group (8.1%). The DC group were experience periodontal disease (33.4%), dental caries (30.1%), diabetes (41.8%), myocardial infarction (57.3%), arthritis (44.0%), asthma (48.0%), and depression (41.9%). In addition, 86% of the DC group were experiencing speech problems. The DC group had significantly lower intakes (1446.59g), than the NDC group (1666.62g), and the protein, carbohydrate, dietary fiber and other dietary intake were significantly lower. These findings suggest that the chewing difficulty is related to the nutrient intake, and psychological status in the elderly DC group. Therefore, the care of chewing difficulties is essential for the elderly to maintain a healthy lifestyle. Accordingly, oral care and myofunctional therapy are needed to maintain oral health.

• Annual Conference on Human and Language Technology
• /
• 2015.10a
• /
• pp.129-132
• /
• 2015

본 논문에서는 단어열 패턴과 리커런트 신경망을 이용한 하이브리드 음성 인식 오류 수정 방법을 제안한다. 음성 인식 결과 문장에서 음성 인식 오류 단어가 발견되었을 경우에 첫째로 단어열 패턴과 그 패턴의 발음열 점수를 통해 1차적 수정을 하고 적절한 패턴을 찾지 못하였을 경우 음절단위로 구성된 Recurrent Neural Network를 통해 단어를 음절단위로 생성하여 2차적으로 오류를 수정한다. 해당 방법론을 한국어로 된 음성 인식 오류와 그 정답 문장으로 구성된 TV 가이드 영역 말뭉치를 바탕으로 성능을 평가하였고, 기존의 단순 단어열 패턴 기반의 음성 인식 오류 수정보다 성능이 향상되었음을 볼 수 있었다. 이 방법론은 음성 인식 오류와 정답의 말뭉치가 필요 없이 옳은 문장으로만 구성된 일반 말뭉치만으로 훈련이 가능하여, 음성 인식 엔진에 의존적이지 않는 강점이 있다.

• The Journal of the Korea Contents Association
• /
• v.14 no.2
• /
• pp.527-534
• /
• 2014

Motivated by the fact that English grapheme is realized as 5 different English vowels [ɑ, eɪ, æ, ɔ, ə], the present study aims to investigate how Korean students perceive and produce English grapheme in words. To this end, 31 university students participated in the perception and production tests on 28 English words containing grapheme . Overall, the participants performed better on production than perception both in the pretest and posttest. Also, the accuracy of perception and production was improved in the posttest. While the accuracy rates of [ɑ] and [eɪ] were among highest in the perception test, those of [ɑ] and [ɔ] were in the production test. By contrast, the accuracy of unstressed [ə] was worst both in perception and production. As for the training effect, [eɪ] and [æ] showed significant improvement both in perception and production. In addition, the perception and the production of [ɔ] were improved at the .051-4 significance level. However, the accuracies of [ɑ] improved significantly neither in perception nor in production. In the case of [ə], the production ability was improved significantly whereas the perception ability was not. The prominent error patterns of were the replacement of the target vowels with [æ], [ɑ], or [eɪ]. The influence of orthography was attested through the persistent occurrence of [ɑ]. Finally, pedagogical implications were discussed.

• Journal of the HCI Society of Korea
• /
• v.11 no.3
• /
• pp.13-22
• /
• 2016

Vocal and instrument apprenticeship in applied music has been accepted without significant changes since the introduction of applied music education to South Korea. Few discussions or suggestions about other types of teaching than 'one-to-one lessons' or 'education of apprentices by assigned specialists' have been made. Since the introduction of applied music education to South Korea late in the 1980s, the CAI(Computer Aided Instruction) courseware development for applied music education has not actively been under way. The area of applied music has also made rapid progress in terms of music producing or music videos using computers. Actually, the improved computer program is not positively applied to applied music education. This study aimed to present learning methods using the improved functions of music production softwares to improve the traditional apprenticeship system in the area of vocal training in applied music. In particular, it used the technique of auto tune-pitch shift developed for interval correction in sound sources. By giving real-time feedbacks concerning intervals or monitors visually after recording, it intended to present a learning method to induce improvement in accuracy of intervals in vocal training. This study is expected to present a method that allows vocal trainers to overcome temporal and spatial limitations in applied music and make their vocal training more efficient.

• Phonetics and Speech Sciences
• /
• v.11 no.1
• /
• pp.41-49
• /
• 2019

An end-to-end speech recognition model consisting of a single integrated neural network model was recently proposed. The end-to-end model does not need several training steps, and its structure is easy to understand. However, it is difficult to understand how the model recognizes speech internally. In this paper, we visualized and analyzed the attention-based end-to-end model to elucidate its internal mechanisms. We compared the acoustic model of the BLSTM-HMM hybrid model with the encoder of the end-to-end model, and visualized them using t-SNE to examine the difference between neural network layers. As a result, we were able to delineate the difference between the acoustic model and the end-to-end model encoder. Additionally, we analyzed the decoder of the end-to-end model from a language model perspective. Finally, we found that improving end-to-end model decoder is necessary to yield higher performance.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.32 no.2
• /
• pp.229-235
• /
• 2005

A problem that affects children's dentitions is the harmful habit which is difficult to treat. Harmful habits for children are such as abnormal swallowing patterns, low/forward tongue rest posture problem, habitual open-lips resting posture, habitual mouth-breathing, excessive digital sucking habit and tongue thrusting. Tongue thrusting habits cause a bit of cranio-facial skeletal changes and a great deal of dental malocclusion such as anterior open bite. Anterior open bite causes masticatory, speech, and esthetic problems in the growing children and difficulties in diagnosis, treatment, and the prediction of its prognosis. The treatments of such abnormal behaviors involve orofacial myofunctional therapy and using of habit breaking appliance. The prognosis is not determined by the presence of severity of oral habit but the skeletal tendency of the patient. Usage of tongue crib resulted in not only the discontinuance of the habit but also improvement in overbite and overbite. This study showed that relatively successful results could be generated by using removable tongue crib and myofunctional therapy in the case of openbite related to tongue thrusting habit.

• Journal of Korea Game Society
• /
• v.7 no.2
• /
• pp.83-90
• /
• 2007

One result of the trend towards globalization is an increased number of projects that focus on natural language processing. Automatic speech recognition (ASR) technologies, for example, hold great promise in facilitating global communications and collaborations. Unfortunately, to date, most research projects focus on single widely spoken languages. Therefore, the cost to adapt a particular ASR tool for use with other languages is often prohibitive. This work takes a more general approach. We propose an International Phoneticizing Engine (IPE) that interprets input files supplied in our Phonetic Language Identity (PLI) format to build a dictionary. IPE is language independent and rule based. It operates by decomposing the dictionary creation process into a set of well-defined steps. These steps reduce rule conflicts, allow for rule creation by people without linguistics training, and optimize run-time efficiency. Dictionaries created by the IPE can be used with the speech recognition system. IPE defines an easy-to-use systematic approach that can obtained 92.55% for the recognition rate of Korean speech and 89.93% for English.

• Journal of Naturopathy
• /
• v.9 no.2
• /
• pp.62-70
• /
• 2020

Purpose: According to the Body-EU education program, this study investigates changes after conducting education with ten subjects with autism symptoms (disability level 2~3) 63 times (126 hours) for nearly two years. Methods: Body-EU habit practice, note writings discussion, singing, Bible study, question and communication, class arrangement, etc., a total of 63 times were educated for 2 hours each week. Results: Subjects felt comfortable breathing and improved energy. Other subject's habit of an out-toed gait step and bending back alleviated. One subject made it easier to speak, so if talking with no strength. When breathing, the stuffy nose improved, the eyes became calm, and the hands and feet were warm. Subject felt like my head was cleaned. Heart tightness disappeared. Subject felt at ease as gaining confidence. The subject gained confidence in seeing whoever the other person was. The subject's pronunciation became more sophisticated, with more confidence, and his mouth cramping alleviated. Subjects with right eye movements showed a relatively high concentration in education time. Conclusions: Subjects are educated and trained in making general habits of their body from morning to going to bed. This study was a research program for ordinary people, but it confirmed with autism. Whether it was genetic autism or acquired autism, they could be somewhat relieved.