• Phonetics and Speech Sciences
• /
• v.16 no.1
• /
• pp.67-76
• /
• 2024

Recent advances in text-to-speech (TTS) technology have significantly improved the quality of synthesized speech, reaching a level where it can closely imitate natural human speech. Especially, TTS models offering various voice characteristics and personalized speech, are widely utilized in fields such as artificial intelligence (AI) tutors, advertising, and video dubbing. Accordingly, in this paper, we propose a one-shot multi-speaker TTS system that can ensure acoustic diversity and synthesize personalized voice by generating speech using unseen target speakers' utterances. The proposed model integrates a speaker encoder into a TTS model consisting of the FastSpeech2 acoustic model and the HiFi-GAN vocoder. The speaker encoder, based on the pre-trained RawNet3, extracts speaker-specific voice features. Furthermore, the proposed approach not only includes an English one-shot multi-speaker TTS but also introduces a Korean one-shot multi-speaker TTS. We evaluate naturalness and speaker similarity of the generated speech using objective and subjective metrics. In the subjective evaluation, the proposed Korean one-shot multi-speaker TTS obtained naturalness mean opinion score (NMOS) of 3.36 and similarity MOS (SMOS) of 3.16. The objective evaluation of the proposed English and Korean one-shot multi-speaker TTS showed a prediction MOS (P-MOS) of 2.54 and 3.74, respectively. These results indicate that the performance of our proposed model is improved over the baseline models in terms of both naturalness and speaker similarity.

• The Journal of Industrial Distribution & Business
• /
• v.10 no.8
• /
• pp.33-44
• /
• 2019

Purpose - As the scope of existing digital transformation expanded to various degrees, the Fourth Industrial Revolution came into being. In 2016, Klaus Schwab, Chairman of the World Economic Forum (WEF), said that the new technologies that lead the fourth industrial revolution are AI, Block chain, IoT, Big Data, Augmented Reality, and Virtual Reality. This technology is expected to be a full-fledged fusion of digital, biological and physical boundaries. Everything in the world is connected to the online network, and the trend of 'block chain' technology is getting attention because it is a core technology for realizing a super connective society. If the block chain is commercialized at the World Knowledge Forum (WKF), it will be a platform that can be applied to the entire industry. The block chain is rapidly evolving around the financial sector, and the impact of block chains on logistics, medical services, and public services has increased beyond the financial sector. Research design, data, and methodology - Figure analysis of data and social science analytical software of IBM SPSS AMOS 23.0 and IBM Statistics 23.0 were used for all the data researched. Data were collected from hotel employees in China from 25th March to 10th May. Results - The purpose of this study is to investigate the effect of the block chain characteristics of the existing hotel reservation system on the intention to use and to examine the influence of the block chain characteristics of the hotel reservation system on the intention to use, We rearranged the variables having the same or similar meaning and analyzed the effect of these factors on the intention to use the block chain characteristic of the hotel reservation system. 339 questionnaires were used for analysis. Conclusions - There are only sample hotel workers in this study, and their ages are in their 20s and 30s. In future studies, samples should be constructed in various layers and studied. In this study, the block chain characteristics are set as five variables as security, reliability, economical efficiency, availability, and diversity. Among them, Security and reliability made positive effects on the perceived usefulness. Also, security and economics did on the perceived ease. Availability and diversity did on both perceived usefulness and perceived ease. Perceived ease did on perceived usefulness. And perceived ease and perceived usefulness did on user intent. But security and economics did not on the perceived usefulness

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.45 no.11
• /
• pp.1638-1648
• /
• 2016

This study was conducted to analyze the quality characteristics of kimchi in terms of garlic content (0~4.5%). Kimchi was made at $4^{\circ}C$ for 8 weeks, and pH, acidity, organic acid content, free sugar content, microbial counts, flavor pattern, and sensory characteristics were measured. The results show that kimchi containing garlic had a higher pH and lower acidity during fermentation than control kimchi without garlic. Principal component analysis enabled differentiation of the flavor pattern of kimchi according to fermentation period and garlic content. Addition of garlic to kimchi significantly decreased the numbers of total bacteria and lactic acid bacteria for 2 weeks after production. The numbers of total bacteria and lactic acid bacteria increased rapidly up to 2 weeks during fermentation and thereafter decreased gradually. Coliform counts were higher in the control than in kimchi containing garlic, whereas there was no detection after 4 weeks. Yeast and mold counts decreased significantly with increasing garlic content during the initial fermentation stage. Counts could not be detected in kimchi containing garlic. After 4 weeks, counts could not be detected in kimchi without garlic. Among kimchi with different garlic contents, fermentation was slower in kimchi with high garlic content; scores for off-odor and off-note taste were lower as well.

• The Journal of the Petrological Society of Korea
• /
• v.6 no.2
• /
• pp.77-85
• /
• 1997

The Early Proterozoic reworked rock association occurs within the Preacmbrian high grade metamorphic rocks in the area of Daqingshan, Inner Molgolia. In this association, the various large scale ductile deformation belts, form a nappe structure where the foliation steeply dips to north and the lineation ($340^{circ}-30^{\circ}$) plunges at $45^{\circ}55^{\circ}$. This result indicates the subduction/extension with northern part thrusting over the southern part at high angle. The southern subducted microlithon has the characteristics of prograde metamorphism. The northern thrusted microlithon shows the evidence of retrograde metamorphism with decreasing pressure and increasing temperature. The main rock types of Early Proterozoic Moyites are biotite adamellite and syenogranites occurring in the form of small batholiths or stocks and alkali-feldspar granites in veins. The biotite adamellites are progressively contacted with the Archean and Early Proterozoic rocks and contain a great deal of enclaves of metamorphosed rocks, suggesting an anatexis origin. The geochemical characteristics of moyites show the typical features of anatexis granite. At middle to late Early Proterozoic time, the continent-continent collision formed the large scale thrusting and imbrication of Archean basement rocks. According to the mineral assemblage and thermobarometer of Paria et al. (1988) give the following P-T condition : up-faulted block; $700-710^{\circ}C$, 0.72-0.78 Gpa (early stage) and $600^{\circ}C$, 0.44 Gpa (late stage), footwall block; $620^{\circ}C$, 0.8 Gpa (early stage), $620-840^{\circ}C$, 0.64-0.45 Gpa (peak) and $620-630^{\circ}C$, 0.35Gpa (late stage). These results suggest a clockwise P-T-t path (jin et al., 1991, 1994). According to the depth-temperature model in the comperature subduction zone and the experimental data of Wyllie et al. (1983), we propose a tectonic-magmatic-thermal model to account for metamorphism-anatexis of moyite occurring in subduction-shear zone.

• Sleep Medicine and Psychophysiology
• /
• v.27 no.2
• /
• pp.56-66
• /
• 2020

Objectives: In this study, the clinical characteristics of OSA patients and the quality of life before and after CPAP use were compared to determine the degree of improvement in quality of life according to CPAP use. Methods: Age, sex, height, weight, body mass index, Epworth Sleepiness Scale, Modified Mallampatti Score, Montreal Cognitive Assessment-Korean, and Pittsburgh Sleep Quality Index were compared between men and women through medical records. To understand the degree of improvement in quality of life resulting from use of CPAP, a personal telephone call was made to compare the VAS scores for quality of life before and after CPAP use. Results: In height (HT) (Z = -4.525, p < 0.001), weight (BW) (Z = -2.844, p < 0.05), sleep quality (PSQI) (Z = -2.671, p < 0.05), and arousal index (AI) (Z = -2.105, p < 0.05), there was a difference between men and women (p < 0.05). There was no difference in the remaining variables. Cross-analysis (Chi-square test) confirmed a difference between severity and sex of OSA. It has been found that there is no statistically significant order in size according to level-specific severity of OSA for PreCPAP QOL, PostCPAP QOL, CPAPUse Months, and CPAP4Hr/d (%) (p > 0.05). The difference between AHI before and after CPAP was 36.48 ± 21.54 (t = 11.609, p < 0.001) and the difference between QOL before and after CPAP was -25.43 ± 22.06 (t = -7.901, p < 0.001), both of which were significant (p < 0.001). Conclusion: Among OSA patients, there were differences in height (HT), weight (BW), sleep quality (PSQI), arousal index (AI), and severity of OSA between men and women, but the quality of life before and after CPAP was different. However, there was no difference between men and women in quality of life before and after CPAP. In addition, quality of life in OSA patients improved after using CPAP.

• Journal of Korean Society of Coastal and Ocean Engineers
• /
• v.33 no.6
• /
• pp.265-274
• /
• 2021

Outlier detection research in ocean data has traditionally been performed using statistical and distance-based machine learning algorithms. Recently, AI-based methods have received a lot of attention and so-called supervised learning methods that require classification information for data are mainly used. This supervised learning method requires a lot of time and costs because classification information (label) must be manually designated for all data required for learning. In this study, an autoencoder based on unsupervised learning was applied as an outlier detection to overcome this problem. For the experiment, two experiments were designed: one is univariate learning, in which only SST data was used among the observation data of Deokjeok Island and the other is multivariate learning, in which SST, air temperature, wind direction, wind speed, air pressure, and humidity were used. Period of data is 25 years from 1996 to 2020, and a pre-processing considering the characteristics of ocean data was applied to the data. An outlier detection of actual SST data was tried with a learned univariate and multivariate autoencoder. We tried to detect outliers in real SST data using trained univariate and multivariate autoencoders. To compare model performance, various outlier detection methods were applied to synthetic data with artificially inserted errors. As a result of quantitatively evaluating the performance of these methods, the multivariate/univariate accuracy was about 96%/91%, respectively, indicating that the multivariate autoencoder had better outlier detection performance. Outlier detection using an unsupervised learning-based autoencoder is expected to be used in various ways in that it can reduce subjective classification errors and cost and time required for data labeling.

• Journal of Intelligence and Information Systems
• /
• v.24 no.4
• /
• pp.111-136
• /
• 2018

In this paper, we propose a methodology to extract answer information about queries from various types of unstructured documents collected from multi-sources existing on web in order to expand knowledge base. The proposed methodology is divided into the following steps. 1) Collect relevant documents from Wikipedia, Naver encyclopedia, and Naver news sources for "subject-predicate" separated queries and classify the proper documents. 2) Determine whether the sentence is suitable for extracting information and derive the confidence. 3) Based on the predicate feature, extract the information in the proper sentence and derive the overall confidence of the information extraction result. In order to evaluate the performance of the information extraction system, we selected 400 queries from the artificial intelligence speaker of SK-Telecom. Compared with the baseline model, it is confirmed that it shows higher performance index than the existing model. The contribution of this study is that we develop a sequence tagging model based on bi-directional LSTM-CRF using the predicate feature of the query, with this we developed a robust model that can maintain high recall performance even in various types of unstructured documents collected from multiple sources. The problem of information extraction for knowledge base extension should take into account heterogeneous characteristics of source-specific document types. The proposed methodology proved to extract information effectively from various types of unstructured documents compared to the baseline model. There is a limitation in previous research that the performance is poor when extracting information about the document type that is different from the training data. In addition, this study can prevent unnecessary information extraction attempts from the documents that do not include the answer information through the process for predicting the suitability of information extraction of documents and sentences before the information extraction step. It is meaningful that we provided a method that precision performance can be maintained even in actual web environment. The information extraction problem for the knowledge base expansion has the characteristic that it can not guarantee whether the document includes the correct answer because it is aimed at the unstructured document existing in the real web. When the question answering is performed on a real web, previous machine reading comprehension studies has a limitation that it shows a low level of precision because it frequently attempts to extract an answer even in a document in which there is no correct answer. The policy that predicts the suitability of document and sentence information extraction is meaningful in that it contributes to maintaining the performance of information extraction even in real web environment. The limitations of this study and future research directions are as follows. First, it is a problem related to data preprocessing. In this study, the unit of knowledge extraction is classified through the morphological analysis based on the open source Konlpy python package, and the information extraction result can be improperly performed because morphological analysis is not performed properly. To enhance the performance of information extraction results, it is necessary to develop an advanced morpheme analyzer. Second, it is a problem of entity ambiguity. The information extraction system of this study can not distinguish the same name that has different intention. If several people with the same name appear in the news, the system may not extract information about the intended query. In future research, it is necessary to take measures to identify the person with the same name. Third, it is a problem of evaluation query data. In this study, we selected 400 of user queries collected from SK Telecom 's interactive artificial intelligent speaker to evaluate the performance of the information extraction system. n this study, we developed evaluation data set using 800 documents (400 questions * 7 articles per question (1 Wikipedia, 3 Naver encyclopedia, 3 Naver news) by judging whether a correct answer is included or not. To ensure the external validity of the study, it is desirable to use more queries to determine the performance of the system. This is a costly activity that must be done manually. Future research needs to evaluate the system for more queries. It is also necessary to develop a Korean benchmark data set of information extraction system for queries from multi-source web documents to build an environment that can evaluate the results more objectively.

• Culinary science and hospitality research
• /
• v.22 no.4
• /
• pp.24-36
• /
• 2016

This study investigated the quality characteristics of Haengbyung using dry rice flour added with various amounts (0, 20, 40, 60%) of raw and steamed apricot juice. The moisture contents, color values, pH, texture and sensory evaluation(attribute difference and acceptance) for the samples were tested. The moisture content (p<0.001) and pH(p<0.001) of the Haengbyung significantly decreased with the increase of apricot juice. As for the color of the Haengbyung samples, the higher percentage of the apricot juice showed the lower L-value (p<0.001), and higher a-value(p<0.001) and b-value(p<0.001). Texture profile analysis of Haengbyung samples showed that hardness(p<0.001) and gumminess(p<0.001) increased, while adhesiveness(p<0.001) and cohesiveness(p<0.05) decreased with increasing amounts of apricot juice. An attribute difference test of Haengbyung samples showed that appearance(apricot color(p<0.001) and dryness(p<0.001)), flavor(sour(p<0.001), acerbic (p<0.001), and green fruit(p<0.001)), and taste(sour(p<0.001), acerbic(p<0.001), and astringent(p<0.05) as increased as apricot juice amounts increased. The control and Haengbyung sample with 20% of apricot juice (both raw and steamed) showed the highest acceptance in taste, texture, and overall acceptability in consumer ranking testing.

• Journal of Korean Academy of Nursing Administration
• /
• v.7 no.3
• /
• pp.455-472
• /
• 2001

The purpose of this study was to serve as a basis for mapping out successful strategies to build the professional self-image of nurses through uniform, by examining their satisfaction with uniform and professional self-image by general characteristics, and identifying the correlational relationship between the two factors. The subjects in this study were 205 nurses who served in general departments of two university hospitals in Seoul. The departments where special uniform was required, including ICU, RR, OR and CSR, were excluded. The data were collected from September 15 through 22, 2000. For measurement of uniform satisfaction level, a uniform satisfaction scale was prepared by myself, and Arthur(1990)'s PSCNI translated and modified by Song Kyong Ae and Rho Chun Hee(1996) was employed. The Cronbach a of the uniform satisfaction scale and PSCNI was 0.90 and 0.86 respectively. The collected data were analyzed by SAS, and real number, percentage, average and standard deviation were calculated. Besides, t-test, one-way ANOVA, Pearson's r procedures were utilized, and Scheffe test was conducted as a posttest. The findings of this study were as below: 1. The uniform satisfaction of the nurses investigated was scored 2.52 on the basis of 4 points, which was on the medium level. By subarea, symbolicity satisfaction was 2.48, and aesthetic satisfaction was 2.60. The functionability satisfaction was 2.44. So the esthetic satisfaction was greatest. 2. Among general characteristics of the subjects, two factors made a significant difference to their uniform satisfaction : age(F=4.05, P=.0189), and total career(F=4.25, P=.0061). 3. Their professional self-image got 2.75 on the basis of 4 points, which was on upper middle level. The subarea score was 2.79 for professional work, 2.52 for satisfaction and 2.97 for communication. The communication area was rated highest, and the satisfaction area was scored lowest. 4. Among the general characteristics of the subjects, professional self-image was different according to five factors : age(F=17.83, P=.001), marital status(T=5.18, P=.0000), educational background(F=8.72, P=.0002), position(T=-5.29, P=.0000) and total career(F=15.23, p=.0001). Better professional self-image was possessed by the older group than the younger one, by the married group than the singles, by the better-educated group than the less-educated, by the nurses in position equal to or higher than charge nurse, or by the higher-career group. 5. The correlational relationship of uniform satisfaction to professional self-image was statistically significant, yet very weak(r=.1978, p=.0045). The satisfaction area of professional self-image was correlated to every uniform satisfaction area, including symbolicity(4=.4393, p=.0001), aesthetics(r=.2471, p=.0004), functionability(r=.3094, p=.0001) and total satisfaction(r=.4050, p=.0001). Therefore, the uniform satisfaction gave an impact on the satisfaction area of professional self-image of the nurses, and there was a significant correlational relationship between uniform symbolicity area and total professional self-image(r=.2416, p=.0005).

• Journal of the Korean Society of Food Science and Nutrition
• /
• v.42 no.12
• /
• pp.2004-2011
• /
• 2013

This study was carried out to investigate the chemical and microbiological properties of Kimchi containing different amounts of squid and octopus during fermentation at $4^{\circ}C$ for 56 days. Kimchi were assigned to one the of the following 7 treatment groups: CK (control Kimchi; no addition), SK-5 (added 5% of squid), SK-10 (added 10% of squid), SK-20 (added 20% of squid), OK-5 (added 5% of octopus), OK-10 (added 10% of octopus), and OK-20 (added 20% of octopus). The quality characteristics of CK and other treatments were determined by measuring the pH, titratable acidity, reducing sugar, volatile basic nitrogen (VBN), and microbial analysis. During fermentation, Kimchi containing squid and octopus had higher pH and lower acidity values than the CK group (P<0.05). The titratable acidity was increased in all treatment groups, whereas the reducing sugar content decreased gradually during fermentation. The reducing sugar contents were highest in the CK group. The VBN value increased in all treatment groups with increasing storage period, and the CK group had lower VBN value than the other treatment groups containing squid and octopus. The numbers of total bacteria and lactic acid bacteria rapidly increased up to the second week, and this slowly increased thereafter.