• MALSORI
• /
• no.53
• /
• pp.75-92
• /
• 2005

Prosody Generation Based on C-ToBI Representation for Text-to-SpeechSeungwon Kim, Yu Zheng, Gary Geunbae Lee, Byeongchang KimProsody modeling is critical in developing text-to-speech (TTS) systems where speech synthesis is used to automatically generate natural speech. In this paper, we present a prosody generation architecture based on Chinese Tone and Break Index (C-ToBI) representation. ToBI is a multi-tier representation system based on linguistic knowledge to transcribe events in an utterance. The TTS system which adopts ToBI as an intermediate representation is known to exhibit higher flexibility, modularity and domain/task portability compared with the direct prosody generation TTS systems. However, the cost of corpus preparation is very expensive for practical-level performance because the ToBI labeled corpus has been manually constructed by many prosody experts and normally requires a large amount of data for accurate statistical prosody modeling. This paper proposes a new method which transcribes the C-ToBI labels automatically in Chinese speech. We model Chinese prosody generation as a classification problem and apply conditional Maximum Entropy (ME) classification to this problem. We empirically verify the usefulness of various natural language and phonology features to make well-integrated features for ME framework.

• The Bulletin of The Korean Astronomical Society
• /
• v.46 no.1
• /
• pp.41.1-41.1
• /
• 2021

We generate solar magnetograms and EUV images from Galileo sunspot drawings using a deep learning model based on conditional generative adversarial networks. We train the model using pairs of sunspot drawing from Mount Wilson Observatory (MWO) and their corresponding magnetogram (or UV/EUV images) from 2011 to 2015 except for every June and December by the SDO (Solar Dynamic Observatory) satellite. We evaluate the model by comparing pairs of actual magnetogram (or UV/EUV images) and the corresponding AI-generated one in June and December. Our results show that bipolar structures of the AI-generated magnetograms are consistent with those of the original ones and their unsigned magnetic fluxes (or intensities) are well consistent with those of the original ones. Applying this model to the Galileo sunspot drawings in 1612, we generate HMI-like magnetograms and AIA-like EUV images of the sunspots. We hope that the EUV intensities can be used for estimating solar EUV irradiance at long-term historical times.

• Journal of the Korea Convergence Society
• /
• v.5 no.4
• /
• pp.101-106
• /
• 2014

Most peoples are used to prefer to view the video contents rather than the other contents since the video contents are more easy to understand with both their eyes and ears. As the wide spread use of smartphones, the demands for the contents services are increasing rapidly. To promote the contents business, it's important to provide security of subscriber authentication and corresponding communication channels through which the contents are delivered. Generally, symmetric key encryption scheme is used to protect the contents in the channel, and the session key should be upadated periodically for the security reasons. In addition, to protect viewing paid contents by illegal users, the proxy authentication should not be allowed. In this paper, we propose biometric based user authentication and one time key generation models. The proposed model is consist of biometric template registration, session key generation and chanel encryption steps. We analyze the difference and benefits of our model with existing CAS models which are made for CATV contents protection, and also provides applications of our model in electronic commerce area.

• 제어로봇시스템학회:학술대회논문집
• /
• 2005.06a
• /
• pp.427-432
• /
• 2005

This paper is concerned with a simultaneous design of TSK (Takagi-Sugeno-Kang)-linguistic fuzzy models with uncertain model output and the computationally efficient representation. For this purpose, we use the fundamental idea of linguistic models introduced by Pedrycz and develop their comprehensive design framework. The design process consists of several main phases such as (a) the automatic generation of the linguistic contexts by probabilistic distribution using CDF (conditional density function) and PDF (probability density function) (b) performing context-based fuzzy clustering preserving homogeneity based on the concept of fuzzy granulation (c) augment of bias term to compensate bias error (d) combination of TSK and linguistic context in the consequent part. Finally, we contrast the performance of the enhanced models with other fuzzy models for automobile MPG predication data and coagulant dosing process in a water purification plant.

• Journal of the Korea Society of Computer and Information
• /
• v.24 no.3
• /
• pp.41-47
• /
• 2019

In this paper, we propose a application of conditional generative adversarial network (cGAN) for generation of contrast enhanced computed tomography (CT) image. Two types of CT data which were the enhanced and non-enhanced were used and applied by the histogram equalization for adjusting image intensities. In order to validate the generation of contrast enhanced CT data, the structural similarity index measurement (SSIM) was performed. Prepared generated contrast CT data were analyzed the statistical analysis using paired sample t-test. In order to apply the optimized algorithm for the lymph node cancer, they were calculated by short to long axis ratio (S/L) method. In the case of the model trained with CT data and their histogram equalized SSIM were $0.905{\pm}0.048$ and $0.908{\pm}0.047$. The tumor S/L of generated contrast enhanced CT data were validated similar to the ground truth when they were compared to scanned contrast enhanced CT data. It is expected that advantages of Generated contrast enhanced CT data based on deep learning are a cost-effective and less radiation exposure as well as further anatomical information with non-enhanced CT data.

• Smart Media Journal
• /
• v.10 no.4
• /
• pp.28-34
• /
• 2021

Drape is one of the factors that determine the shape of clothes and is one of the very important factors in the textile and fashion industry. At a time when non-face-to-face transactions are being activated due to the impact of the coronavirus, more and more companies are asking for drape value. However, in the case of small and medium-sized enterprises (SMEs), it is difficult to measure the drape, because they feel the burden of time and money for measuring the drape. Therefore, this study aimed to generate a drape image for the material property value input using a conditional adversarial neural network through 3D simulation images generated by measuring digital properties. A drape image was created through the existing 736 digital property values, and this was used for model training. Then, the drape value was calculated for the image samples obtained through the generative model. As a result of comparing the actual drape experimental value and the generated drape value, it was confirmed that the error of the peak number was 0.75, and the average error of the drape value was 7.875

• BMB Reports
• /
• v.42 no.6
• /
• pp.315-323
• /
• 2009

In order to elucidate ultimate biological function of the genome, the model animal system carrying mutations is indispensable. Recently, large-scale mutagenesis projects have been launched in various species. Especially, the mouse is considered to be an ideal model to human because it is a mammalian species accompanied with well-established genetic as well as embryonic technologies. In 1990', large-scale mouse mutagenesis projects firstly initiated with a potent chemical mutagen, N-ethyl-N-nitrosourea (ENU) by the phenotype-driven approach or forward genetics. The knockout mouse mutagenesis projects with trapping/conditional mutagenesis have then followed as Phase II since 2006 by the gene-driven approach or reverse genetics. Recently, the next-generation gene targeting system has also become available to the research community, which allows us to establish and analyze mutant mice carrying an allelic series of base substitutions in target genes as another reverse genetics. Overall trends in the large-scale mouse mutagenesis will be reviewed in this article particularly focusing on the new advancement of the next-generation gene targeting system. The drastic expansion of the mutant mouse resources altogether will enhance the systematic understanding of the life. The construction of the mutant mouse resources developed by the forward and reverse genetic mutagenesis is just the beginning of the annotation of mammalian genome. They provide basic infrastructure to understand the molecular mechanism of the gene and genome and will contribute to not only basic researches but also applied sciences such as human disease modelling, genomic medicine and personalized medicine.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.11
• /
• pp.1456-1462
• /
• 2016

In this paper, we apply a chord progression learning model to a rule-based generation of a four-part chorus. The proposed system is given a 32-note melody line and completes the four-part chorus based on the rule of harmonics, predicting the chord progression with the CRBM model. The data for the training model was collected from various harmony textbooks, and chord progressions were extracted with key-independent features so as to utilize the given data effectively. It was shown that the output piece obtained with the proposed learning model had a more natural progression than the piece that used only the rule-based approach.

• Journal of KIISE:Software and Applications
• /
• v.37 no.11
• /
• pp.837-844
• /
• 2010

Prediction of the prosodic phrase boundary is one of the most important natural language processing tasks. We propose, for the natural prediction of the Korean prosodic phrase boundary, a statistical approach incorporating efficient learning features. These new features reflect the factors that affect generation of the prosodic phrase boundary better than existing learning features. Notably, moreover, such learning features, extracted according to the hand-crafted prosodic phrase boundary prediction rule, impart higher accuracy. We developed a statistical model for Korean prosodic phrase boundaries based on the proposed new features. The results were 86.63% accuracy for three levels (major break, minor break, no break) and 81.14% accuracy for six levels (major break with falling tone/rising tone, minor break with falling tone/rising tone/middle tone, no break).

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.9
• /
• pp.381-386
• /
• 2023

Many existing test case generation researchers extract test cases from models. However, research on generating test cases from natural language requirements is required in practice. For this purpose, the combination of natural language analysis and requirements engineering is very necessary. However, Requirements analysis written in Korean is difficult due to the diverse meaning of sentence expressions. We research test case generation through natural language requirement definition analysis, C3Tree model, cause-effect graph, and decision table steps as one of the test case generation methods from Korean natural requirements. As an intermediate step, this paper generates test cases from C3Tree model-based decision tables using meta-modeling. This method has the advantage of being able to easily maintain the model-to-model and model-to-text transformation processes by modifying only the transformation rules. If an existing model is modified or a new model is added, only the model transformation rules can be maintained without changing the program algorithm. As a result of the evaluation, all combinations for the decision table were automatically generated as test cases.