• Journal of Digital Contents Society
• /
• v.18 no.6
• /
• pp.1101-1107
• /
• 2017

This study explored the effect of ASR-based pronunciation instruction, using a mobile platform, on EFL learners' pronunciation development. Particularly, this quasi-experimental study focused on whether using mobile ASR, which provides voice-to-text feedback, can enhance the perception and production of target English consonants minimal pairs (V-B, R-L, and G-Z) of Korean EFL learners. Three intact classes of 117 Korean university students were assigned to three groups: a) ASR Group: ASR-based pronunciation instruction providing textual feedback by the mobile ASR; b) Conventional Group: conventional face-to-face pronunciation instruction providing individual oral feedback by the instructor; and the c) Hybrid Group: ASR-based pronunciation instruction plus conventional pronunciation instruction. The ANCOVA results showed that the adjusted mean score for pronunciation production post-test on the Hybrid instruction group (M=82.71, SD =3.3) was significantly higher than the Conventional group (M=62.6, SD =4.05) (p<.05).

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.91-94
• /
• 1998

우리만은 영어와는 달리 단어를 공백으로만 구분할 수 없다. 그러므로 대용량 어휘를 갖는 연속 음성을 인식하기 위한 언어모델을 만들기가 매우 어렵다. N-gram의 언어 모델을 우리말 문장에 적용하기 위해 하나의 문장을 한 단어로 구성하여 처리하였다. 우리의 인식시스템을 평가하기 위하여 시스템 공학 연구소에서 제공한 음성을 대상으로 인식률을 계산하였다. 단어의 종류는 452개이며 한명이 이 단어들을 2번씩 발음하고 총70명이 발음한 총 63,280개의 단어에 대하여 92.8%의 인식률을 얻었다. 일간지 사설로부터 추출한 단어를 대상으로 발음 사전을 10K 크기로 만들었다. 음성 모델은 uniphone을 사용하였다.

• The Journal of the Acoustical Society of Korea
• /
• v.18 no.5
• /
• pp.21-27
• /
• 1999

In this paper, we propose an improved MLP-based English pronunciation dictionary generator to apply to the variable vocabulary word recognizer. The variable vocabulary word recognizer can process any words specified in Korean word lexicon dynamically determined according to the current recognition task. To extend the ability of the system to task for English words, it is necessary to build a pronunciation dictionary generator to be able to process words not included in a predefined lexicon, such as proper nouns. In order to build the English pronunciation dictionary generator, we use context-dependent grapheme-to-phoneme multi-layer perceptron(MLP) architecture for each grapheme. To train each MLP, it is necessary to obtain grapheme-to-phoneme training data from general pronunciation dictionary. To automate the process, we use dynamic programming(DP) algorithm with some distance metrics. For training and testing the grapheme-to-phoneme MLPs, we use general English pronunciation dictionary with about 110 thousand words. With 26 MLPs each having 30 to 50 hidden nodes and the exception grapheme lexicon, we obtained the word accuracy of 72.8% for the 110 thousand words superior to rule-based method showing the word accuracy of 24.0%.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.193-196
• /
• 2000

본 논문에서는 제한 지속시간 모델링(bounded duration modeling)의 지속시간 제한 값(duration bound)을 화자의 발음 속도에 따라 조절해주는 발음 속도 보상 알고리즘을 제안한다. 제안된 알고리즘은 두 번의 인식 과정을 수행하는데, 1차 인식 과정은 화자의 발음 속도를 추정하기 위한 과정이고, 2차 인식 과정이 인식 결과를 얻기 위한 과정이다. 1차 인식 과정에서 추정된 화자의 발음 속도에 따라, 지속시간 제한 값을 증가, 또는 감소시킨 후, 2차 인식과정에 사용한다. 제안된 알고리즘은 CHMM 기반의 한국어 연결 숫자음 인식 시스템에 적용되었으며, KAIST에서 제작된 4-7자리 연결 숫자음 데이터베이스인 DigitDB를 대상으로 성능을 평가하였다. 인식 실험 결과, 제안된 발음 속도 보상 알고리즘이 적용된 인식 시스템에서는 $96.26\% 단어 인식률을 얻어, 제안된 알고리즘이 적용되지 않은 인식 시스템의 $94.72\%보다 $1.54\% 향상된 인식 성능을 얻을 수 있었다.

• The Journal of Korean Academy of Prosthodontics
• /
• v.60 no.1
• /
• pp.37-43
• /
• 2022

Patients tend to return to normal pronunciation patterns after fitting new dentures. However, for some patients, it takes a long time to adapt the new complete denture. In this case, the patient came to the hospital at the address of wanting to remake dentures due to wear and tear. After diagnosis through clinical and radiological examination, the maxillary complete denture and mandibular removable partial denture were remade. The patient complained whistling /s/ sound at the first check-up after placement of the new denture. The anterior palatal area of polished surface of the new maxillary complete denture was concave comparing to old denture, and this was the cause of the whistling /s/ sound. A tissue conditioning material was applied to the maxillary complete denture and patient made /s/ sound. The tissue conditioning material was replaced with self-curing type denture base resin, and the patient was immediately satisfied with clear /s/ sound. As an objective assessment, palatogram and speech analytics software was applied. In this case, a patient who received denture treatment complaining of difficulty in pronunciation underwent immediate denture repair, which resulted in patient satisfaction and improved pronunciation through objective evaluation.

• Journal of Yeungnam Medical Science
• /
• v.7 no.2
• /
• pp.49-54
• /
• 1990

Ten cleft palate patients were operated with reconstruction of levator sling without pushback for the purpose of not to make raw surface in the anterior portion of hard palate to prevent maxillary retrognathia. Speech was evaluated by using speech assessment list. Maxillary growth was not evaluated due to in-growing age in majority patient. The report will be followed in next chance. We could impose the significance in clinical application of levator sling palatoplasty without any complications but improving speech.

• The Journal of the Korea Contents Association
• /
• v.14 no.3
• /
• pp.482-489
• /
• 2014

The purposes of this study are (1) to see if the length of residence in Korea affects to the VOT values of Korean plosives produced by Chinese speakers (2) to investigate if such VOT values are different in non-sense words and real words and (3) to find out how such VOT values are related to the native Korean speakers' understanding of the words. For these purposes, eighteen Chinese speakers' VOT values of Korean plosives were measured using Speech Analyzer and their pronunciations were evaluated by three native Korean speakers. The results show that there is no effect of both (1) length of residence in Korea and (2) production of non-sensed words and sensed words. In addition, it is noticed that VOT values of some plosives affect more to the native Korean speakers' understanding of the words, while some are not.

• Journal of Dental Rehabilitation and Applied Science
• /
• v.39 no.3
• /
• pp.119-132
• /
• 2023

The loss of posterior occlusal support due to tooth loss is likely to lead to compensatory protrusion and labial tilt of the anterior teeth, which may be accompanied by a deep bite and a decrease in vertical dimension. The patient may suffer from a decrease in masticatory efficiency, inaccurate pronunciation, facial appearance changes, and temporomandibular joint disorder, so stable occlusal formation with support of posterior occlusion and restoration of vertical dimension is necessary. We report the case of a patient with reduction of vertical dimension, and inaccurate pronunciation due to multiple tooth loss who underwent full mouth rehabilitation with increased vertical dimension accompanied by phonetic analysis and achieved satisfactory functional and aesthetic results.

• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06c
• /
• pp.87-90
• /
• 1998

이 논문에서 우리는 대규모 어휘를 갖는 연속 음성 인식을 위한 방법을 제시한다. 우리말은 영어와 구조적으로 달라서 대용량 어휘를 갖는 연속 음성을 인식하기 위한 언어모델을 만들기가 매우 어렵다. 언어 모델을 우리말 문장에 적용하기 위해 신문의 사설을 3-gram을 이용하여 처리하였다. 우리의 인식 시스템을 평가하기 위하여 시스템 공학 연구소에서 제공한 낭독 음성을 대상으로 인식률을 계산하였다. 589개의 문장을 대상으로 총 20명이 발음한 3,156개의 문장에 대하여 남자 92.2%, 여자 87.9%의 인식률을 얻었다. 발음사전은 낭독음성과 신문 사설에서 추출한 10K 크기이며 uniphone의 음성모델을 사용하였다.

• Proceedings of the Korean Society for Emotion and Sensibility Conference
• /
• 2001.11a
• /
• pp.49-54
• /
• 2001

본 연구에서는 가전제품의 동작상태를 사용자에게 적절히 알릴 수 있는 정보음의 적합성 평가기준을 마련하기 위하여 선행연구 결과와 주부들을 대상으로 가전제품의 사용실태를 분석하였으며, 고령자들도 알아듣기 쉽고 생활소음에 mask 되지 않으며 들어서 쾌적하도록 1, 1.6, 2, 3, 4 kHz의 완전 5도화음으로 구성된 정보음을 제작하여, 평균발음빈도(F)가 1.67 Hz인 그룹과 0.91 Hz인 그룹의 Attac(A)과 Decay(D)를 변화시켜 A/D의 변화에 따른 단순반응시간을 측정한 결과 F가 1.67Hz 그룹에서 A/D가 작을수록 단순반응시간이 짧아지는 경향을 확인하였으며, F가 1.67Hz와 0.91Hz의 수준에 따른 단순반응시간 차이검증 결과 p < 0.001 수준에서 유의미하여 5점 척도로 평가하였으며, 요인분석결과 첫 번째 요인은 긴박성요인, 두 번째 요인은 명료성 요인, 세 번째 요인은 역능요인, 네 번째 요인은 평가요인으로 나타났다.