• Phonetics and Speech Sciences
• /
• v.11 no.2
• /
• pp.57-64
• /
• 2019

In this paper, we construct a Korean text-to-speech system using the Merlin toolkit which is an open source system for speech synthesis. In the text-to-speech system, the HMM-based statistical parametric speech synthesis method is widely used, but it is known that the quality of synthesized speech is degraded due to limitations of the acoustic modeling scheme that includes context factors. In this paper, we propose an acoustic modeling architecture that uses deep neural network technique, which shows excellent performance in various fields. Fully connected deep feedforward neural network (DNN), recurrent neural network (RNN), gated recurrent unit (GRU), long short-term memory (LSTM), bidirectional LSTM (BLSTM) are included in the architecture. Experimental results have shown that the performance is improved by including sequence modeling in the architecture, and the architecture with LSTM or BLSTM shows the best performance. It has been also found that inclusion of delta and delta-delta components in the acoustic feature parameters is advantageous for performance improvement.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.2
• /
• pp.187-192
• /
• 2022

Even the announcing date for the staring date of "With Corona" has been decided, still many people have not completed vaccination, the most important condition for starting the With Corona, because of concerns for its side effects. In addition, although the economy may can be recovered by the With Corona, but the number of infected people may can be surged. In this paper, in order to awaken the people for the awareness of Corona 19 in advance of the With Corona, the Corona 19 is predicted through a non-linear probability process. Here, among the deep learning RNN, BI-LSTM, which is a bidirectional LSTM, and GRU, gates decreased than LSTM have been used. And this has been compared and analyzed through train set, test set, loss function, residual analysis, normal distribution, and autocorrelation, and compared and predicted for which has a better performance.

• Food Science of Animal Resources
• /
• v.43 no.4
• /
• pp.612-624
• /
• 2023

The gut-brain axis encompasses a bidirectional communication pathway between the gastrointestinal microbiota and the central nervous system. There is some evidence to suggest that probiotics may have a positive effect on cognitive function, but more research is needed before any definitive conclusions can be drawn. Inflammation-induced by lipopolysaccharide (LPS) may affect cognitive function. To confirm the effect of probiotics on oxidative stress induced by LPS, the relative expression of antioxidant factors was confirmed, and it was revealed that the administration of probiotics had a positive effect on the expression of antioxidant-related factors. After oral administration of probiotics to mice, an intentional inflammatory response was induced through LPS i.p., and the effect on cognition was confirmed by the Morris water maze test, nitric oxide (NO) assay, and interleukin (IL)-1β enzyme-linked immunosorbent assay performed. Experimental results, levels of NO and IL-1β in the blood of LPS i.p. mice were significantly decreased, and cognitive evaluation using the Morris water maze test showed significant values in the latency and target quadrant percentages in the group that received probiotics. This proves that intake of these probiotics improves cognitive impairment and memory loss through anti-inflammatory and antioxidant mechanisms.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.12 no.4
• /
• pp.456-465
• /
• 2019

PHEV(Plug in Hybrid Electric Vehicle) and BEV(Battery Electric Vehicle) equip high voltage batteries to drive motor and vehicle electric system. Those vehicle require OBC(On-Board Charger) for charging batteries and LDC(Low DC/DC Converter) for converting from high voltage to low voltage. Since the charger and the converter actually separate each other in electrical vehicles, there is a margin to reduce the vehicle weight and area of installation by integration two systems. This paper studies a 1.5kW LDC converter that can be integrated into an OBC using an isolated current-fed converter by simplifying the design of LDC transformers. The proposed LDC can control the final output voltage of the LDC by using a fixed arbitrary output voltage of the bidirectional buck-boost converter, so that Compared to the existing OBC-LDC integrated system, it has the advantage of simplifying the transformer design considering the battery voltage range, converter duty ratio and OBC output turn ratio. Prototype of the proposed LDC was made to confirm normal operation at 200V ~ 400V input voltage and maximum efficiency of 91.885% was achieved at rated load condition. In addition, the OBC-LDC integrated system achieved a volume of about 6.51L and reduced the space by 15.6% compared to the existing independent system.

• Journal of Advanced Navigation Technology
• /
• v.28 no.5
• /
• pp.760-766
• /
• 2024

Machine learning and deep learning are being researched in various fields and applied in real life. Designing reliable models is crucial, and understanding the results of these models is necessary. This paper analyzes the impact of variables on prediction values using SHAP. Prediction models for PM_2.5 were designed using DNN and LSTM algorithms. The training and test data were composed by selecting weather data and air pollutant data through correlation analysis. The RMSE and accuracy for AQI categories were checked for both prediction models, with the LSTM algorithm showing slightly better performance. The contribution of variables to the prediction values of both models was confirmed using SHAP. It was found that air pollutant data had a high contribution in predicting PM_2.5, and temperature among weather data had a high contribution in the prediction process of both models. Both models showed that high values of temperature, wind speed, and sea level pressure decreased prediction values, while low values increased them. For NO₂ , PM₁₀, and SO₂, the LSTM model showed a bidirectional impact on prediction values, unlike the DNN model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.6
• /
• pp.1154-1162
• /
• 2016

This paper proposes a fast-transient digital LDO(Low dropout) regulator with binary-weighted current control technique. Conventional digital LDO takes a long time to stabilize the output voltage, because it controls the amount of current step by step, thus ringing problem is generated. Binary-weighted current control technique rapidly stabilizes output voltage by removing the ringing problem. When output voltage reliably reaches the target voltage, It added the FRZ mode(Freeze) to stop the operation of digital LDO. The proposed fast response digital LDO is used with a slow response DC-DC converter in the system which rapidly changes output voltage. The proposed digital controller circuit area was reduced by 56% compared to conventional bidirectional shift register, and the ripple voltage was reduced by 87%. A chip was implemented with a $0.18{\mu}F$ CMOS process. The settling time is $3.1{\mu}F$ and the voltage ripple is 6.2mV when $1{\mu}F$ output capacitor is used.

• Information Systems Review
• /
• v.15 no.3
• /
• pp.71-87
• /
• 2013

Companies as well as the Korean government show growing attention to the Smart Work which is enabled by the ongoing development of information and communication technologies (ICTs). Smart Work can be regarded as an extended version of telecommuting or distance work and defined as "working efficiently and conveniently regardless of time and place utilizing ICTs." Smart Work currently puts its emphasis on the work and life balance by changing the ways of working. Despite its emphasis on work and life balance, it is expected the boundaries between work and nonwork would blur, work and nonwork boundaries may become more permeable, and role conflicts would occur more times than before. To find ways to enhance work and life balance while escaping from expected conflicts in the context of Smart Work, we investigate the work and nonwork boundary strengths and the factors affecting them. In the course, we consider asymmetries between the work and nonwork boundary strengths and bidirectional nature of work and nonwork permeability. We develop two research models having work and nonwork boundary strengths as respective dependent variables. We empirically found that work role identification and nonwork-to-work permeability had influences on the boundary strength at work and that work-to-nonwork permeability affected the boundary strength at nonwork. However, nonwork role identification did not show any significant influence on the boundary strength at nonwork.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.10
• /
• pp.709-721
• /
• 2006

The traditional broadcasting services over terrestrial, satellite and cable media have been unidirectional mass media regardless of TV viewer's preferences. Recently ich media streaming has become possible via the broadb and networks. Furthermore, since bidirectional communication is possible, personalcasting such as personalized streaming services has been emerging by taking into account the user's preference on content genres, viewing times and actors/actresses etc. Accordingly, personal media becomes an important means for content provision service in addition to the traditional broadcasting service as mass media. In this paper, we introduce a user profile reasoning method for target advertisement which is considered an important application in personalcasting service. The proposed user profile reasoning method predicts an unknown TV viewer's gender and ages by analyzing TV Viewing history data. Based on the estimated user's gender and ages, a target advertisement is provided with TV Anytime metadata. A proposed target advertisement system is developed based on the user profile reasoning and the target advertisement selection method. To show the effectiveness of our proposed methods, we present a plenty of experimental results by using realistic TV viewing history data.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.2B
• /
• pp.302-312
• /
• 2010

Internet Protocol Television (IPTV) service is the convergence service of the telecommunication and broadcasting that provides various bidirectional multimedia contents by IPTV service subscribe's request through the high-speed internet. However, the proposed technologies current do not guarantee the security such as authentication between Set-Top-Box (STB) and the user mobile devices available IPTV service at home domain, and authentication of mobile user device at out of door. This paper proposes the authentication protocol for distributing content securely from STB to the users' mobile devices at home domain and authentication for network access and IPTV service access when the user's mobile device is moved out of the house. The proposed scheme using the proxy signature enables to distribute and protect securely the contents protected through an underlying Conditional Access System (CAS) without re-encrypting then that the existing scheme should employ. Then this protocol supports the authentication scheme to get service access authentication based on network access authentication using the signature, which the STB issued on behalf of the trust authority of IPTV service provider. Also the proposed authentication protocol reduces the total communication overhead and computation time comparing to the other authentication protocol.

• Management & Information Systems Review
• /
• v.37 no.2
• /
• pp.63-85
• /
• 2018

Electronic displays have developed as instruments that function as windows of information to satisfy the bidirectional needs of electronic device manufacturers and users. The cathode-ray tube (CRT) display once dominated the display industry. Now, the liquid crystal display (LCD) is dominant, and promising displays are competing to become the next-generation display. Displays reflect the characteristics of the information and communication technology (ICT) industries, such as technological changes, innovative features, and competitive dynamics, and have technologically evolved to dominate the industry and market through various ICT devices. This research utilizes a dominant design concept and examines the case of the flat-panel display industry to propose a comprehensive framework, which considers technological, organizational, and environmental characteristics, of the determinants influencing dominance in the technology-intensive ICT industry. The results show that a dominant design in the flat-panel display industry emerges from technological competition among several designs, based on technological characteristics and market acceptance, and dominates the industry and market by various environmental factors. Our results emphasize the difference between generic technologies and the speed of technological innovation and expand the understanding of the emergence of a dominant design. Furthermore, this paper suggests practical implications for establishing a competition strategy and strategic guidance for other ICT industries as well as the display industry.