• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.536-543
• /
• 2023

Speaker diarization, which labels for "who spoken when?" in speech with multiple speakers, has been studied on a deep neural network-based end-to-end method for labeling on speech overlap and optimization of speaker diarization models. Most deep neural network-based end-to-end speaker diarization systems perform multi-label classification problem that predicts the labels of all speakers spoken in each frame of speech. However, the performance of the multi-label-based model varies greatly depending on what the threshold is set to. In this paper, it is studied a speaker diarization system using single-label classification so that speaker diarization can be performed without thresholds. The proposed model estimate labels from the output of the model by converting speaker labels into a single label. To consider speaker label permutations in the training, the proposed model is used a combination of Permutation Invariant Training (PIT) loss and cross-entropy loss. In addition, how to add the residual connection structures to model is studied for effective learning of speaker diarization models with deep structures. The experiment used the Librispech database to generate and use simulated noise data for two speakers. When compared with the proposed method and baseline model using the Diarization Error Rate (DER) performance the proposed method can be labeling without threshold, and it has improved performance by about 20.7 %.

• Journal of the Earthquake Engineering Society of Korea
• /
• v.10 no.6 s.52
• /
• pp.115-123
• /
• 2006

This paper presents long-term dynamic characteristics of a cable-stayed bridge where installed SHM (Structural Health Monitoring) system. Modal parameters such as natural frequencies and mode shapes are identified by modal analysis using three dimensional finite element model. The developed baseline model has a good correlation with measured natural frequencies identified from field ambient vibrations. By statistical data processing between measured natural frequencies and temperatures, it is demonstrated that the natural frequency is in linearly inverse proportion to the temperature. The estimation of temperature effects against frequency variations is performed. Mode shapes are identified from the TDD (Time Domain Decomposition) technique for ambient vibration measurements. Finally, these results demonstrate that the TDD method can apply to identify modal parameters of a cable-stayed bridge.

• Korean Journal of Environment and Ecology
• /
• v.28 no.3
• /
• pp.374-380
• /
• 2014

This study was conducted to provide baseline information which will be useful for managing the forest trails as the outdoor recreation facility. To investigate residents' attitudes toward the forest trail construction and use, a questionnaire survey was employed in Jirisan Dulegil. As the results, 4 impact factors that are influenced to local communities due to Jirisan Dulegil construction were resulted in. Most of the respondents perceived that the Jirisan Dulegil have positive impact economically on community growth. Contrastively, a negative perception existed in the environmental aspect. Local lodging industry was selected one of preference business models. In viewpoint of community business, a more professional and systematic approach to find appropriate business model is needed because of structural vulnerability of business.

• Journal of the Korean Dietetic Association
• /
• v.21 no.3
• /
• pp.215-226
• /
• 2015

The prevalence of obesity is increasing worldwide and has become a serious epidemic health problem. We developed the 'Change 10 Habits' educational program based on obesity treatment and dietary guidelines and examined its effects on customized nutrition education in mildly obese adults. The study was approved by the Institutional Review Board. Study subjects were excluded if they had several major diseases, if had consumed an anti-obesity drug, or if they practiced an obesity-related program within 30 days. The subjects (n=87, $25{\leq}BMI$ <30) were each exposed to the customized nutrition education program with four lessons according to the stage of the transtheoretical model (TTM). The stage-matched program was administered for 12 weeks and was run by a clinical dietitian. Overall, subjects who were in the precontemplation/contemplation stage at baseline made progress in the preparation and action/maintenance stage after 12 weeks (P<0.05). For 'Alcohol is consumed, up to 2 drinks per day', the proportion of subjects who belonged in the action/maintenance stage increased from 34.5% to 49.4% at 12 weeks. In addition, scores of all items significantly increased after the program (P<0.05). 'Chew more than 10 times and eat slowly' score significantly increased from $3.9{\pm}2.4$ to $5.8{\pm}2.3$ (P<0.05). In conclusion, behavioral stage-matched nutrition education using the 'Change 10 Habits' program was effective in improving eating behaviors and enhancing healthy lifestyles in mildly obese adults.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.41 no.7
• /
• pp.516-523
• /
• 2013

The tip clearance effect on counter-rotating ducted fan of VTOL UAV in hovering condition, was investigate using computational analysis. The $k-{\omega}$ SST turbulence model is employed in this study. The numerical results of baseline model are validated by wind tunnel test in hovering and forward conditions. It is observed that if tip clearance of one rotor in the counter-rotating ducted fan increase then the thrust coefficient of another rotor increases. In Addition to this, when the tip clearance of the rear rotor increases, the thrust of the ducted fan is improved due to increasing of average total pressure at exit plane.

• Korean Journal of Clinical Laboratory Science
• /
• v.45 no.4
• /
• pp.131-138
• /
• 2013

Insulin resistance and pancreatic beta cell dysfunction have been established as being related to the diabetes. Lately, what is emphasizing is that those have been shown as something related to the metabolic syndrome and cardiovascular disease. Homeostasis model assessment (HOMA), simple index is calculated on blood levels of fasting glucose and insulin. And HOMA has been widely validated and applied for insulin resistance and pancreatic beta cell dysfunction. We also assessed the factors relative to insulin resistance and ${\beta}$ cell function determined by HOMA. The data from the 2010 Korean National Health and Nutrition Examination Survey were used. Analysis was done for 3,465 nondiabetic subjects (male 1,357, female 2,108). At baseline, anthropometric measurements were done and fasting glucose, insulin, lipid (Total cholesterol, HDL cholesterol, LDL cholesterol and Triglycerides) profiles were measured. HOMA-insulin resistance (HOMA-IR) and beta cell function (HOMA ${\beta}$-cell) were calculated from fasting glucose and insulin levels. In male, the value of HOMA-IR and HOMA ${\beta}$-cell was the highest among 30's and decreased as the age increased. In female, the value of HOMA-IR increased with age, while HOMA ${\beta}$-cell decreased. High HOMA-IR and low HOMA ${\beta}$-cell were associated with the highest value of fasting glucose and systolic blood pressure. Low HOMA-IR and high HOMA ${\beta}$-cell showed the lowest concentration of fasting glucose and the highest concentration of HDL cholesterol. High HOMA-IR and high HOMA ${\beta}$-cell were connected with BMI, Total cholesterol, LDL cholesterol, and Triglycerides. There was a negative correlation between HOMA ${\beta}$-cell and age. The correlation coefficients of HOMA-IR and HOMA ${\beta}$-cell showed the highest value among weight, BMI and WC.

• Korean Journal of Construction Engineering and Management
• /
• v.3 no.4 s.12
• /
• pp.114-122
• /
• 2002

Earned Value Management System(EVMS) is a management technique that is emerging as a valuable tool in the management of all projects, including and in particular construction projects. It is based on C/SCSC that has been released by the U.S. Department of Defense (DOD) in 1967. In a project that employs the earned value management concept, the project's measurable performance plans will be formed with the creation of CAPs. The formation of CAPs is a planning step necessary to form an earned value baseline. In this paper, a method is proposed to form CAPs using Material Requirement Planning(MRP) that is also known as a useful planning tool in the manufacturing industry. An application to a real project has been carried through workpackaging model.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.36 no.4
• /
• pp.667-675
• /
• 2016

The goal of this study is to be used as baseline data for transportation demand management. At the present time the number of registered car and householding car is increasing, so there is a need to analyze the car ownership pattern through household car hold status. Also, it is necessary to analyze the factor of increasing car. The research is proceeded with classifying as the household which is holding private cars or holding passenger cars and non passenger cars based on the result of the research of the household travel survey data. The result of this study is shown as follows. According to car ownership pattern, there are more households holding passenger cars only when they are holding less than 2 cars. Otherwise there are more households holding passenger car and non passenger car when they are holding more than 3 cars. Using the Ordered Logit Model, there are more differences in factors affects holding cars by variables of housing type and household properties.

• The Korean Journal of Pain
• /
• v.23 no.3
• /
• pp.186-189
• /
• 2010

Background: Neuropathic pain resulting from diverse causes is a chronic condition for which effective treatment is lacking. The goal of this study was to test whether dexamethasone exerts a preemptive analgesic effect with bupivacaine when injected perineurally in the spared nerve injury model. Methods: Fifty rats were randomly divided into five groups. Group 1 (control) was ligated but received no drugs. Group 2 was perineurally infiltrated (tibial and common peroneal nerves) with 0.4% bupivacaine (0.2 ml) and dexamethasone (0.8 mg) 10 minutes before surgery. Group 3 was infiltrated with 0.4% bupivacaine (0.2 ml) and dexamethasone (0.8 mg) after surgery. Group 4 was infiltrated with normal saline (0.2 ml) and dexamethasone (0.8 mg) 10 minutes before surgery. Group 5 was infiltrated with only 0.4% bupivacaine (0.2 ml) before surgery. Rat paw withdrawal thresholds were measured using the von Frey hair test before surgery as a baseline measurement and on postoperative days 3, 6, 9, 12, 15, 18 and 21. Results: In the group injected preoperatively with dexamethasone and bupivacaine, mechanical allodynia did not develop and mechanical threshold forces were significantly different compared with other groups, especially between postoperative days 3 and 9 (P < 0.05). Conclusions: In conclusion, preoperative infiltration of both dexamethasone and bupivacaine showed a significantly better analgesic effect than did infiltration of bupivacaine or dexamethasone alone in the spared nerve injury model, especially early on after surgery.

• Journal of KIISE
• /
• v.43 no.6
• /
• pp.678-685
• /
• 2016

Named Entity Recognition and Classification (NERC) is a task for recognition and classification of named entities such as a person's name, location, and organization. There have been various studies carried out on Korean NERC, but they have some problems, for example lacking some features as compared with English NERC. In this paper, we propose a method that uses word embedding as features for Korean NERC. We generate a word vector using a Continuous-Bag-of-Word (CBOW) model from POS-tagged corpus, and a word cluster symbol using a K-means algorithm from a word vector. We use the word vector and word cluster symbol as word embedding features in Conditional Random Fields (CRFs). From the result of the experiment, performance improved 1.17%, 0.61% and 1.19% respectively for TV domain, Sports domain and IT domain over the baseline system. Showing better performance than other NERC systems, we demonstrate the effectiveness and efficiency of the proposed method.