• ETRI Journal
• /
• 제32권5호
• /
• pp.810-818
• /
• 2010

This paper presents a new framework for integrating untranscribed spoken content into the acoustic training of an automatic speech recognition system. Untranscribed spoken content plays a very important role for under-resourced languages because the production of manually transcribed speech databases still represents a very expensive and time-consuming task. We proposed two new methods as part of the training framework. The first method focuses on combining initial acoustic models using a data-driven metric. The second method proposes an improved acoustic training procedure based on unsupervised transcriptions, in which word endings were modified by broad phonetic classes. The training framework was applied to baseline acoustic models using untranscribed spoken content from parliamentary debates. We include three types of acoustic models in the evaluation: baseline, reference content, and framework content models. The best overall result of 18.02% word error rate was achieved with the third type. This result demonstrates statistically significant improvement over the baseline and reference acoustic models.

• 한국건축시공학회:학술대회논문집
• /
• 한국건축시공학회 2015년도 추계 학술논문 발표대회
• /
• pp.198-199
• /
• 2015

In Korea, it raised the need of the specification of unit modular construction, because there are no standard or manual. And it is urgent that the specification of unit modular construction is needed to develop in the domestic. Therefore, this study aims to propose the manual framework of the construction method. In the future, this framework will be based on the development of the manual on the unit modular construction method.

• 대한전기학회논문지:전력기술부문A
• /
• 제48권10호
• /
• pp.1222-1230
• /
• 1999

This paper presents the framework for analysis of small-signal stability. In this framework the equilibrium points of system DAE model are traced using continuation method and instability points are determined on the solution path. Especially Implicit shift QR-modified Arnoldi method is utilized to calculated the rightmost eigenvalues. Small-signal stability is becoming one of the major factors that affect the ATC in the environment of deregulation. Stability limited ATC is easily evaluated in this stability analysis process.

• 한국BIM학회 논문집
• /
• 제12권4호
• /
• pp.52-60
• /
• 2022

Recently, the demand for connection between various heterogeneous dataset and BIM as a construction data model hub is increasing. In the past, in order to connect model between BIM and heterogeneous dataset, related dataset was stored in the RDBMS, and the service was provided by programming a method to link with the BIM object. This approach causes problems such as the need to modify the database schema and business logic, and the migration of existing data when requirements change. This problem adversely affects the scalability, reusability, and maintainability of model information. This study proposes an ontology BIM-based knowledge service framework considering the connectivity and scalability between BIM and heterogeneous dataset. Through the proposed framework, ontology BIM mapping, semantic information query method for linking between knowledge-expressing dataset and BIM are presented. In addition, to identify the effectiveness of the proposed method, the prototype is developed. Also, the effectiveness and considerations of the ontology BIM-based knowledge service framework are derived.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권8호
• /
• pp.2720-2735
• /
• 2022

This study proposes an internet design framework for users to improve their positive emotions when they are in a negative mood. First, the literature review focuses on the definition of emotion, positive emotional design in internet experiences, and emotion regulation. Second, in order to construct an internet design framework that improves positive emotion, this paper adopts a qualitative analysis method to analyze 70 collected studies in the area of regulating emotion and stimulating positive emotions. Additionally, bibliometrics and statistics are conducted to summarize the framework and strategies. Third, two cases of internet design are presented: (a) Internet design that improves users' positive emotions is examined under the background of extreme rainstorm as an example; an applet service design is provided by case study; (b) in the context of COVID-19, we developed an Internet of things interactive design that improves users' positive emotions. Fourth, the internet design framework and the results of the case studies are analyzed and discussed. Finally, an internet design framework is proposed to improve users' positive emotions when they are in a negative mood, which includes the Detachment-empathy framework, External-protection framework, Ability-strengthen framework, Perspective-transformation framework, and Macro-cognitive framework. The framework can help designers to generate design ideas accurately and quickly when users are in a negative mood, to improve subjective well-being, and contribute to the development of internet experience design.

• 대한음성학회지:말소리
• /
• 제45호
• /
• pp.93-105
• /
• 2003

In this paper, we present a single framework based on GMM for speaker recognition. The proposed framework can simultaneously minimize environmental variations on mismatched conditions and adapt the bias free and speaker-dependent characteristics of claimant utterances to the background GMM to create a speaker model. We compare the closed-set speaker identification for conventional method and the proposed method both on TIMIT and NTIMIT. In the several sets of experiments we show the improved recognition rates on a simulated channel and a telephone channel condition by 7.2% and 27.4% respectively.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1998년도 추계학술대회 논문집 학회본부A
• /
• pp.201-203
• /
• 1998

This paper presents the framework for analysis of small-signal stabili1ty. In this framework the equilibrium points of system DAE model are traced using continuation method and instability points are determined on the solution path. Especially Implicit shift QR-modified ARnoldi method is utilized to calculated the rightmost eigenvalues. ISPS measure is utilized to determine critical parameters.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제29권1_2호
• /
• pp.65-79
• /
• 2002

객체지향 프레임웍은 다수의 응용 소프트웨어의 개발에 반복적으로 재사용되므로 철저한 시험이 요구될 뿐만 아니라 재사용 시 확장된 프레임웍에 대해서도 추가적인 시험이 필요하다. 그런데 프레임웍은 개조, 합성된 확장 부위의 시험에 대한 제어와 관찰을 어렵게 하는 성질을 갖고 있다. 본 논문에서는 프레임웍을 개조, 확장하여 응용 프로그램을 구현할 때 발생할 수 있는 오류들이 시험을 통하여 효율적으로 발견될 수 있도록 프레임웍의 가변 부위에 테스터 컴포넌트들을 BIT(Built-in Test)로 내장하는 방법을 기술한다. 프레임웍에 이와 같이 내장된 테스터 컴포넌트들은 프레임웍의 시험 시 제어와 관찰을 용이하게 하여 프레임웍의 시험성을 높여준다. 여기서 제안된 방법으로 설계된 테스터 컴포넌트들은 시험대상 프레임웍의 확장 부위에 프레임웍 코드의 변경이나 간섭 효과가 없게 부착할 수 있고 필요에 따라 동적으로 탈착할 수 있다.

• 한국경영과학회:학술대회논문집
• /
• 한국경영과학회 2003년도 추계학술대회 및 정기총회
• /
• pp.148-151
• /
• 2003

Recently school food service operations are confronted with the wide spread pressures for accountability and the need to increase productivity. This paper is concerned with the make-or-buy decision framework for school food service systems considering the multi-attributes in the decision making. For the purpose of considering the multi-attributes analysis method in decision making for the school foodservice, we developed a make-or-buy decision framework using the multi-attribute analysis method, analytic hierarchy process, AHP method for school food service system. Finally, we developed a systematic and practical solution builder for a three-step decision support system in the view of 1) brainstorming for the idea generation, 2) analytic hierarchy process, AHP as a multi-attribute structure ed analysis method, and 3) aggregation logic model to integrate the results of reviewers. We developed web based program and applied it to a school foodservice problem.

• 한국정보과학회논문지:소프트웨어및응용
• /
• 제41권8호
• /
• pp.545-556
• /
• 2014

최근 각광받고 있는 연구분야 중 하나인 Internet of Things (IoT)은 사물들이 네트워크 상에 연결되어 있는 있는 컴퓨팅 환경으로, 이 환경을 구성하는 여러 IoT 디바이스들이 협업을 통해 사용자에게 유용한 기능을 제공할 수 있는 장점이 있다. IoT 디바이스의 제한된 배터리 수명, 이동성과 같은 동적인 특성으로 인해 IoT 디바이스의 상태는 계속해서 변하게 되며, 이로 인해 IoT 디바이스를 관리하는 데에 어려움이 있다. 본 논문에서는 IoT 디바이스의 동적인 특성에 의한 IoT 디바이스 관리 문제를 도출하고 이를 해결하기 위한 디바이스 관리 프레임워크를 제안한다. 그리고, 디바이스 검색, 디바이스 상태 확인, 디바이스 선택 및 연결, 그리고 디바이스 교체를 포함한 네 가지의 기법을 제시한다. 또한 제시된 기법들을 기반으로 디바이스 관리 프레임워크를 구현하고 실험을 통하여 제안한 기법들의 실효성을 검증한다.