• Phonetics and Speech Sciences
• /
• v.6 no.1
• /
• pp.77-83
• /
• 2014

This paper proposes a speaker and environment clustering method in order to overcome the degradation of the speech recognition performance caused by various noise and speaker characteristics. In this paper, instead of using the distance between Gaussian mixture model (GMM) weight vectors as in the Google's approach, the distance between the adapted mean vectors based on the modified maximum a posteriori (MAP) adaptation is used as a distance measure for vector quantization (VQ) clustering. According to our experiments on the simulation data generated by adding noise to clean speech, the proposed clustering method yields error rate reduction of 10.6% compared with baseline speaker-independent (SI) model, which is slightly better performance than the Google's approach.

• MALSORI
• /
• no.43
• /
• pp.151-161
• /
• 2002

This study presents performance analysis results of speaker adaptation for keyword spotting system. In this paper, we implemented MLLR (Maximum Likelihood Linear Regression) method on our middle size vocabulary keyword spotting system. This system was developed for directory services of universities and colleges. The experimental results show that speaker adaptation reduces the false alarm rate to 1/3 with the preservation of the mis-detection ratio. This improvement is achieved when speaker adaptation is applied to not only keyword models but also non-keyword models.

• MALSORI
• /
• no.43
• /
• pp.127-136
• /
• 2002

The speaker adaptation technique is generally used to reduce the speaker difference in speech recognition. In this work, we focus on the features fitted to a linear regression-based speaker adaptation. These are obtained by feature transformation based on independent component analysis (ICA), and the transformation matrix is learned from a speaker independent training data. When the amount of data is small, however, it is necessary to adjust the ICA-based transformation matrix estimated from a new speaker utterance. To cope with this problem, we propose a smoothing method: through a linear interpolation between the speaker-independent (SI) feature transformation matrix and the speaker-dependent (SD) feature transformation matrix. We observed that the proposed technique is effective to adaptation performance.

• Phonetics and Speech Sciences
• /
• v.1 no.4
• /
• pp.119-125
• /
• 2009

This paper proposes fast speaker adaptation method using artificially distorted speech in telematics terminal under the car noise environment based on eigenspace-based maximum likelihood linear regression (ES-MLLR). The artificially distorted speech is built from adding the various car noise signals collected from a driving car to the speech signal collected from an idling car. Then, in every environment, the transformation matrix is estimated by ES-MLLR using the artificially distorted speech corresponding to the specific noise environment. In test mode, an online model is built by weighted sum of the environment transformation matrices depending on the driving condition. In 3k-word recognition task in the telematics terminal, we achieve a performance superior to ES-MLLR even using the adaptation data collected from the driving condition.

• Journal of the Korean Institute of Telematics and Electronics
• /
• v.26 no.8
• /
• pp.1167-1177
• /
• 1989

In this paper, a generalized adaptive law is proposed which uses a rational function type operator for parameter adjustment. To satisfy the passivity condition of the adaptation block, we introduce a constant feedback gain into the adaptation block. This adaptation scheme is applied to the model reference adaptive control of a continuous-time, linear time-invariant, minimum-phase system whose relative degree is 1. We prove the asymptotic stability of the output error of this adaptive system by hyperstability method. It is shown that by digital computer simulations this law can give a better output error transient response in some cases than the conventional gradient adaptive law. And the output error responses for the several types of the proposed adaptation law are examined in the presence of a kind of unmodeled dynamics.

• Journal of information and communication convergence engineering
• /
• v.4 no.3
• /
• pp.118-122
• /
• 2006

In this paper, a sliding mode control method with uncertainty adaptation is proposed by introducing the virtual state. Because upper bound of the uncertainty is very difficult to know, we estimate this by using the simple adaptation law and design the sliding surface which has dynamic of nominal system. An optimal controller is used by nominal controller. And if initial values of the virtual state are chosen properly, the reaching phase is removed.

• Journal of Technologic Dentistry
• /
• v.41 no.1
• /
• pp.1-7
• /
• 2019

Purpose: The objectives of this study was to evaluate clinical adaptation of dental prostheses printed by 3 dimensional(3D) printing technology. Methods: Ten study models were prepared. Ten specimens of experimental group were printed by 3D printing(3DP group). As a control group, 10 specimens were fabricated by casting method on the same models. Marginal gaps of all specimens were measured to evaluate clinical adaptation. Marginal adaptations were measured using silicone replica technique and measured at 8 sites per specimen. Wilcoxon's signed-ranks test was used for statistical analysis(${\alpha}=0.05$). Results: Means of marginal adaptations were $95.1{\mu}m$ for 3DP group and $75.9{\mu}m$ for CAST group(p < 0.000). Conclusion : However, the mean of the 3DP group was within the clinical tolerance suggested by the previous researchers. Based on this, dental prosthesis fabricated by 3D printing technology is considered to be clinically acceptable.

• Research in Community and Public Health Nursing
• /
• v.13 no.4
• /
• pp.743-756
• /
• 2002

Purpose: This study is a descriptive and comparative study that compares health state and school adaptation between children in divorced family and in normal family. Study results will provide a basic data for the development of an intervention program designed to help children in divorced family adjust to their crisis. Method: The study subjects consisted of 700 children in 4th. 5th or 6th elementary school grade, residing Seoul and Kyunggi regions. Among these subjects, 123 were children with divorced family and 577 were children with normal family. The health status of the subjects was measured by Health Symptom Questionnaire developed by Shin and revised by the investigators. The instrument consisted of 30 items measuring physical and emotional health symptoms. The level of school adaptation of the subjects was measured by School Adaptation Scale developed by Lee, which consisted of 4 dimensions with 20 items what measures relationship with peer students, learning activity, observance of regulation, and participation of school activity. The investigators visited the schools and collected data in the classes using the questionnaire after explaining the purpose and procedures of the study to the children. The data were analyzed by descriptive statistics, $\yen\"{o}2$-test, t-test, ANOVA using the SPSS PC + statistical program. Result: First, the mean health state score in children with normal family (M=11.99) was higher than that in children with divorced family (M=19.15), showing a significant difference (t=-6.51, p=.000) between the two groups, which suggests that children with normal family have better health state than children with divorced family. Second, the mean school adaptation score in children with normal family (M=38.99) was higher than that in children with divorced family (M=26.97), showing a significant difference between the two groups (t=104.07, p=.000), which suggests that the school adaptation of children with normal family is better than that of children with divorced family. Third, in comparison of health state between the two groups by general characteristics. there were significant differences between the two groups in sex. the most contributing factor to health status of the children, school year. birth order, religion, school achievement, amount of monthly pocket money, parents level of formal education, occupation of parents, economic status (p<.05). Forth, in comparison of the level of school adaptation between the two groups by general characteristics, there were significant differences between the two groups in most variables (p<.05), suggesting that children with normal family had better capacity of school adaptation than children with divorced family. Conclusion: As a result. this study showed that the parent's divorce had great influence on children's health status and school adaptation capacity. The implication for nursing is that there is a need to develop supportive interventions for the high-risk children who have decreased health states and school adaptation capacity due to the divorce of their parents. In addition, it is recommended that further studies should be conducted to explore protective factors for the prevention of health and adaptation problems in children.

• The Korean Journal of Emergency Medical Services
• /
• v.14 no.3
• /
• pp.29-40
• /
• 2010

Purpose: We investigated the motivation of selecting major, the satisfaction of major and the view on occupation of emergency medical technology (EMT) students. The results of study will be used to increase the satisfaction of major of university students. Also, it can be used to support decision of major for high school students. Method: We conducted 1,586 surveys from 665 students at six four-year colleges and 921 students at seven three-year colleges. Our research period was from Aug. 27th to Oct. 31st in 2010. The surveys were analyzed by SPSS 18.0 using description statistics, t-test, ANOVA, Scheffe and Pearson correlation coefficient. Result: In our study, the motivation of selecting major is 35.8% of students selected 'good employment prospects', satisfaction of major's $M{\pm}SD$ is $3.15{\pm}.486$, adaptation of major's $M{\pm}SD$ is $3.11{\pm}.472$. The satisfaction of major show difference (t = 4.548 p = .000) by sex, also the adaptation of major show difference (t = 2.279, p = .023) by sex. The satisfaction of major show first grade students higher (F = 3.605, p = .013) than fourth grade students at four-year colleges. If satisfaction of major is high, accumulation evaluation score is high (F = 3.276, p = .011), too. Clinical practice experienced students was higher (t = -2.878, p = .004) than non experienced it satisfaction. In view of occupation, ideal job's factors and actual job's factors a lot of students selected 'aptitude'. Also, there is a statistically significant correlation (r = .618, p = .000) between the satisfaction of major and the adaptation of major. High satisfaction indicates high adaptation of major. Conclusion: In our study satisfaction of major and adaptation of major was very high score. EMT students concern about employment prospects at most. It is inferred that they select job which match with one's aptitude. We can suppose that students select major as a tool for employment by seeing result that a lot of students consider employment prospects at most when they select major. A method to improve the satisfaction and adaptation of major should be developed by realizing problems which occur the dissatisfaction of major. Also, there as on why students conflict between ideal and actual job should be revealed. There as on seems students want stable occupation in unstable job market situation. Therefore, the expansion of job matching aptitude and being stable should be processed. Finally, university should actively support the method that help to finding jobs for student by identifying job preparation of students.

• The Journal of Advanced Prosthodontics
• /
• v.9 no.2
• /
• pp.124-129
• /
• 2017

PURPOSE. This study evaluated the effect of span length on the adaptation of implant-supported cobalt chromium frameworks fabricated by three techniques. MATERIALS AND METHODS. Models with two solid abutment analogs at different inter-abutment distances were digitized using a laboratory scanner. Frameworks of two-, three-, and four-unit fixed prostheses were designed by a computer. Six dots with a diameter of 0.2 mm were preset on the surface of each framework. A total of 54 implant-supported cobalt chromium frameworks were fabricated by milling, selective laser melting (SLM), and cast techniques. The frameworks were scanned and exported as Stereolithography files. Distances between two dots in X, Y, and Z coordinates were measured in both the designed and fabricated frameworks. Marginal gaps between the framework and the abutments were also evaluated by impression replica method. RESULTS. In terms of distance measurement, significant differences were found between three- and four-unit frameworks, as well as between two- and four-unit frameworks prepared by milling technique (P<.05). Significant differences were also noted between two- and three-unit frameworks, as well as between two- and four-unit frameworks prepared by cast technique (P<.05). The milling technique presented smaller differences than the SLM technique, and the SLM technique showed smaller differences than the cast technique at any unit prostheses (P<.05). Evaluation with the impression replica method indicated significant differences among the span lengths for any fabrication method (P<.05), as well as among the fabrication methods at any unit prostheses (P<.05). CONCLUSION. The adaptation of implant-supported cobalt chromium frameworks was affected by the span length and fabrication method.