• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.05a
• /
• pp.397-400
• /
• 2020

본 논문은 영화에 나오는 효과음을 자막으로 생성해주는 자동자막생성을 제안하며, 그의 첫 단계로써 다중 사운드 분류 모델을 제안하였다. 고양이, 강아지, 사람의 음성을 분류하기 위해 사운드 데이터의 특정벡터를 추출한 뒤, 4가지의 기계학습에 적용한 결과 최적모델로 딥러닝이 선정되었다. 전처리 과정 중 주성분 분석의 유무에 따라 정확도는 81.3%와 33.3%로 확연한 차이가 있었으며, 이는 복잡한 특징을 가지는 사운드를 분류하는데 있어 주성분 분석과 넓고 깊은 형태의 신경망이 보다 개선된 분류성과를 가져온 것으로 생각된다.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.31 no.6D
• /
• pp.849-856
• /
• 2011

This paper proposes a method for transforming unstructured text contents of engineering documents, which have complex hierarchical structure of subtitles with various heading symbols, into a semi-structured XML document according to the hierarchical subtitle structure. In order to extract the hierarchical structure from plain text information, this study employed a method of document structure extraction which is an analysis technique of the document structure. In addition, a method for processing enumerative text contents was developed to increase overall accuracy during extraction of the subtitles and construction of a hierarchical subtitle structure. An application module was developed based on the proposed method, and the performance of the module was evaluated with 40 test documents containing structural calculation records of bridges. The first test group of 20 documents related to the superstructure of steel girder bridges as applied in a previous study and they were used to verify the enhanced performance of the proposed method. The test results show that the new module guarantees an increase in accuracy and reliability in comparison with the test results of the previous study. The remaining 20 test documents were used to evaluate the applicability of the method. The final mean value of accuracy exceeded 99%, and the standard deviation was 1.52. The final results demonstrate that the proposed method can be applied to diverse heading symbols in various types of engineering documents to represent the hierarchical subtitle structure in a semi-structured XML document.

• The Journal of the Korea Contents Association
• /
• v.22 no.4
• /
• pp.28-36
• /
• 2022

This study analyzed 284 K-pop song and dance cover short-form videos recreated by Chinese influencers uploaded on TikTok, to explore which reform factors of image similarity, language similarity, the extent of audience participation leading, the extent of lyrics or subtitles translated into Chinese, PPL disclosure, the length of video and the reputation of influencer affected Chinese TikTok audiences' reactions - number of "Likes," "Comments" and "Shares." The results showed that only the "reputation of influencer" was significantly affected the number of "Likes" which estimated as a relatively passive response, but the other factors affected the number of "Comments" and "Shares" significantly which estimated as more active responses. The more an influencer is perceived as not similar to the singer in terms of image the more comments were posted. And the videos expressed in Korean archived more comments and shares than those lyrics or subtitles translated into Chinese. This study is meaningful in that it confirmed the necessity of influencers in the globe diffusion of K-pop, by specifically analyzing the audience's reactions according to the characteristics of UCCs created by local influencers using short-form video platforms.

• Journal of Fashion Business
• /
• v.19 no.5
• /
• pp.48-62
• /
• 2015

The purpose of the study was to investigate how the Classification of the Joseon Dynasty's Beauty characteristics were expressed in a TV documentary focusing ona Re-mediation theory. The methods of study comprised of library research, Internet search, and using TV documentary program case studies. The work this Researcher makeup in the manufacture from the documentary in which the methods the Joseon. Dynasty expressed were selected. The literature, and preceding research, were referred to as a way to help organize the Joseon Dynasty's 'gi-saeng Hwang Jin-Hee', 'woman of royal family', and way of make-up of 'sadae-bu lady'. The TV documentary programs selected were 'MBC special' and '2 parts of channel A documentary special'. First, the improvisation of nature and simultaneity expressed in the Joseon Dynasty's usage of make-up is shown through the interview form reflecting the make-up tools and age direction of the scenes or expert. Second, the interactivity and reality are well seen through the row equivalent in which the model seems to directly use the dressing demonstration of the expert and cosmetics material. Third, the cultural expandability and unexpectedness show through the production of situations which are viewed from the explanation of the narration and letter subtitles and drama.

• Phonetics and Speech Sciences
• /
• v.13 no.3
• /
• pp.65-70
• /
• 2021

This paper proposes a method for designing a large recording script for open domain English speech synthesis. For read-aloud style text, 12 domains and 294 sub-domains were designed using text contained in five different news media publications. For conversational style text, 4 domains and 36 sub-domains were designed using movie subtitles. The final script consists of 43,013 sentences, 27,085 read-aloud style sentences, and 15,928 conversational style sentences, consisting of 549,683 tokens and 38,356 types. The completed script is analyzed using four criteria: word coverage (type coverage and token coverage), high-frequency vocabulary coverage, phonetic coverage (diphone coverage and triphone coverage), and readability. The type coverage of our script reaches 36.86% despite its low token coverage of 2.97%. The high-frequency vocabulary coverage of the script is 73.82%, and the diphone coverage and triphone coverage of the whole script is 86.70% and 38.92%, respectively. The average readability of whole sentences is 9.03. The results of analysis show that the proposed method is effective in producing a large recording script for English speech synthesis, demonstrating good coverage in terms of unique words, high-frequency vocabulary, phonetic units, and readability.

• Journal of Korea Game Society
• /
• v.19 no.1
• /
• pp.73-84
• /
• 2019

This study is intended to suggest ways to improve the quality of the service of subtitles provided for the convenience of viewing for deaf people on e-sports broadcast content and game streamer content. Generally, subtitling files of broadcast content are manually written on air by stenographers, so a delay of 3 to 5 seconds is inevitable compared to the original content. Therefore, the present study proposed the formation of an automatic synchronization calibration system using speech recognition technology. In addition, a content application experiment using this system was conducted, and the final result confirmed that the time of synchronization error of subtitling data could be reduced to less than 1 second.

• Journal of Distribution Science
• /
• v.20 no.7
• /
• pp.107-117
• /
• 2022

Purpose: The scope of forensic investigations serves to identify malicious activities, including leakage of crucial corporate information. The investigations also identify security lapses in available networks. The purpose of the present study is to explore how to block distribution channels to protect illegal leakage in supply chain through digital forensic method. Research design, data and methodology: The present study conducted the qualitative textual analysis and its data collection process entails five steps: identifying and collecting data, determining coding categories, coding the content, checking validity and reliability, and analyzing and presenting the results. This methodology is a significant research method due to its high quality of previous resources. Results: Applying previous literature analysis to the results of this study, the author figured out that there are four solutions as an evidences to block distribution channels, preventing illegal leakage regarding company information. The following subtitles show clear solutions: (1) Communicate with Stakeholders, (2) Preventing and addressing illegal leakage, (3) Victims of Data Breach, (4) Focusing Solely on Technical Teams. Conclusion: There are difficult scenarios that continue to introduce difficult questions surrounding engagement with digital evidence. Consequently, it is important to enhance data handling to provide answers for organizations that suffer due to illegal leakages of sensitive information.

• The Journal of Korean Medicine
• /
• v.43 no.4
• /
• pp.74-88
• /
• 2022

Objectives: The purpose of this study is to design Korean-themed online public education content in Korean medicine using rapid prototyping instructional systems design (RPISD). This study presents cases of developing and converting face-to-face general education programs designed to increase the interest in and understanding of Korean medicine for the public into online programs within a short timeframe. Methods: This qualitative study is design and development research, which used the RPISD model to analyze the available resources utilized in the rapid development of public educational content and propose systematization and optimization measures by analyzing the needs of clients, learners, and the environment. The <Treasured Mirror of Eastern Medicine(DUBG)Open Course> was developed according to the model procedure, which involved needs analysis, development of course materials and manuscript, and storyboard creation and its filming and editing. Usability tests were conducted at all stages, and the opinions of clients, instructors, experts, and instructional designers were accommodated and reflected at each stage. Results: Using the rapid prototyping model, <Treasured Mirror of Eastern Medicine(DUBG)Open Course> was organized into five classes of 20 minutes each. Each class was developed in Korean and included English, Chinese, and Japanese subtitles in addition to Korean under the cooperative instructional design among clients, subject-matter experts, instructional designer and learners. Conclusion: The cooperative instructional design of stakeholders is significant in developing Korean medicine public education content online through extensive interaction and feedback from stakeholders in the early stage of educational content development.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.563-565
• /
• 2021

The most important feature of the recently increasing content providing service is that the amount of content increase over time is very large. Accordingly, the importance of user curation is increasing, and various techniques are used to implement it. In this paper, among the techniques for video recommendation, the analysis technique using voice data and subtitles and the video comparison technique based on keyframe extraction are compared with the results of implementing and applying the video content of real big data. In addition, through the comparison result, a video content environment to which each analysis technique can be applied is proposed.

• Journal of the Korean Society for Library and Information Science
• /
• v.48 no.2
• /
• pp.25-44
• /
• 2014

This research evaluates the accessibility of the mobile applications for the South Korean libraries based on the accessibility guideline from the Ministry of Security and Public Administration. In order to enhance the credibility of the evaluation, this current research covers both the accessibility for the visually impaired and the accessibility for the people without disabilities. The research found four main results. First, we found that only 21 libraries (31%) provide alternative texts. Out of the 21 libraries, only one provide alternative texts across all sections of the mobile applications, including the main page, data search, information assistance, etc. Second, most of the mobile applications provide contents in texts, and the subtitles, sign language, blinking and background music provided as required or recommended standard by the guideline lack correlation. Third, alternative texts, focus movement, accessibility of operating system, button motion control, spacing between control and alarm functions must follow the standard guideline for the people with disabilities to use the mobile applications. Fourth, follow-up research on the development of accessibility standard for library mobile application is necessary in order to enable people with disabilities to freely use the library mobile applications.