• Proceedings of the Acoustical Society of Korea Conference
• /
• 1998.06d
• /
• pp.60-63
• /
• 1998

본 논문에서는 사용자가 별도의 훈련과정 없이 인식대상 어휘를 추가 및 변경이 가능한 가변어휘 인식시스템에 관하여 기술한다. 가변어휘 음성인식에서는 미리 구성된 음소모델을 토대로 인식대상 어휘가 결정되명 발음사전에 의거하여 이들 어휘에 해당하는 음소모델을 연결함으로써 단어모델을 만든다. 사용된 음소모델은 현재 음소의 앞뒤의 음소 context를 고려한 문맥종속형(Context-Dependent)음소모델인 triphone을 사용하였고, 연속확률분포를 가지는 Hidden Markov Model(HMM)기반의 고립단어인식 시스템을 구현하였다. 비교를 위해 문맥 독립형 음소모델인 monophone으로 인식실험을 병행하였다. 개발된 시스템은 음성특징벡터로 MFCC(Mel Frequency Cepstrum Coefficient)를 사용하였으며, test 환경에서 나타나지 않은 unseen triphone 문제를 해결하기 위하여 state-tying 방법중 음성학적 지식에 기반을 둔 tree-based clustering 기법을 도입하였다. 음소모델 훈련에는 ETRI에서 구축한 POW (Phonetically Optimized Words) 음성 데이터베이스(DB)[1]를 사용하였고, 어휘독립인식실험에는 POW DB와 관련없는 22개의 부서명을 50명이 발음한 총 1.100개의 고립단어 부서 DB[2]를 사용하였다. 인식실험결과 문맥독립형 음소모델이 88.6%를 보인데 비해 문맥종속형 음소모델은 96.2%의 더 나은 성능을 보였다.

• Proceedings of the KSPS conference
• /
• 2003.05a
• /
• pp.75-78
• /
• 2003

In this paper, we describe a cross-morpheme pronunciation variation model which is especially useful for constructing morpheme-based pronunciation lexicon for Korean LVCSR. There are a lot of pronunciation variations occurring at morpheme boundaries in continuous speech. Since phonemic context together with morphological category and morpheme boundary information affect Korean pronunciation variations, we have distinguished pronunciation variation rules according to the locations such as within a morpheme, across a morpheme boundary in a compound noun, across a morpheme boundary in an eojeol, and across an eojeol boundary. In 33K-morpheme Korean CSR experiment, an absolute improvement of 1.16％ in WER from the baseline performance of 23.17％ WER is achieved by modeling cross-morpheme pronunciation variations with a context-dependent multiple pronunciation lexicon.

• 제어로봇시스템학회:학술대회논문집
• /
• 1994.10a
• /
• pp.207-212
• /
• 1994

A mathematical model is developed for a batch reactor in which the free radical bulk copolymerization of styrene and acrylonitrile takes place. In this model, we introduce the free volume theory to quantify the diffusion controlled termination and propagation reactions, and develop a model for the chain length dependent termination reaction in the context of the pseudo kinetic rate constant method(PKRCM). The simulation results from this model are found to be in good agreement with experimental data under different copolymerization conditions. The present model can predict both the copolymer composition and the number and weight average molecular weights. These kinetic approaches provide greater insight into the performance of the batch reactor used for the free radical bulk copolymerization of styrene and acrylonitirle.

• Coupled systems mechanics
• /
• v.7 no.6
• /
• pp.669-690
• /
• 2018

This paper deals with numerical modeling of dynamic failure phenomena in rate-sensitive brittle and/or ductile materials. To this end, a two-dimensional continuum viscodamage-embedded discontinuity model, which is based on our previous work (see Do et al. 2017), is developed. More specifically, the pre-peak nonlinear and rate-sensitive hardening response of the material behavior, representing the fracture-process zone creation, is described by a rate-dependent continuum damage model. Meanwhile, an embedded displacement discontinuity model is used to formulate the post-peak response, involving the macro-crack creation accompanied by exponential softening. The numerical implementation in the context of the finite element method exploiting the second-order mid-point scheme is discussed in detail. In order to show the performance of the model several numerical examples are included.

• Journal of Information Technology Services
• /
• v.12 no.3
• /
• pp.39-54
• /
• 2013

IT consulting is becoming a norm rather than exception in this age of smart work and information revolution. As IT consulting is one of the knowledge intensive services requiring high credence on both sides, maintaining a good trustful relationship is critical in sustenance of strategic partnership between business firms and IT service firms. Trust is known to be one of the salient constructs in service relationships. In this study, building from the social psychology literature, trust is conceptualized as two dimensions : cognitive and affective trust. Using two dimensions of trust as mediators, a research model is constructed for IT consulting specific context : relationship continuance intention as the dependent construct while expertise, service performance, reputation, relationship satisfaction and value similarity as antecedents of cognitive and affective trust. 145 data points were collected through a survey of IT service client project managers retrospectively asking their experience with IT consultants. Findings suggest that cognitive trust is associated with perceived level of expertise and service performance while affective trust with relationship satisfaction and value similarity, respectively. Interestingly, the paths from reputation are found to be statistically insignificant towards both dimensions of trust, indicating IT service context would be more practically outcome oriented than any other professional service context. Also, cognitive trust seems to maintain stronger influence on relationship continuance intention as anticipated. Implications and limitations are discussed at the end.

• MALSORI
• /
• no.45
• /
• pp.117-126
• /
• 2003

Decision tree-based state tying has been proposed in recent years as the most popular approach for clustering the states of context-dependent hidden Markov model-based speech recognition. The aims of state tying is to reduce the number of free parameters and predict state probability distributions of unseen models. But, when doing state tying, the size of a decision tree is very important for word independent recognition. In this paper, we try to construct optimized decision tree based on the average of feature vectors in state pool and the number of seen modes. We observed that the proposed optimal decision tree is effective in predicting the state probability distribution of unseen models.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.7
• /
• pp.519-529
• /
• 2020

Livelihood capitals have a clear influence on livelihood development. As for the livelihood results, it has been pointed out in the analysis of the poor households that the ability of people to escape poverty depends especially on the access to livelihood capitals. This study aims to analyze the impacts of livelihood capital on poverty among mountain people who depend on forests through human capital, social capital, natural capital, physical capital and financial capital. This research employs the model of binary regression function. Independent variables x1, x2, …, xn are targets of livelihood strategy, vulnerability context, and livelihood capitals. These variables were selected to be included in the original model with dependent variable Y as poor and non-poor households. This study surveys households living in upland areas, near forests, and households of ethnic minorities. The results show that,out of the poor household rate, nearly 4% are newly-poor households or those falling back into poverty. Therefore, the government needs to pay more attention to this disadvantaged group and implements policies such as education and training policies, credit support policies, policies to support forest development, and payment for forest environmental services in the context of emerging countries like Vietnam.

• Language and Information
• /
• v.14 no.1
• /
• pp.1-24
• /
• 2010

The complexity of comparative constructions in each language has given challenges to both theoretical and computational analyses. This paper first identifies types of comparative constructions in Korean and discusses their main grammatical properties. It then builds a syntactic parser couched upon the typed feature structure grammar, HPSG and proposes a context-dependent interpretation for the comparison. To check the feasibility of the proposed analysis, we have implemented the grammar into the existing Korean Resource Grammar. The results show us that the grammar we have developed here is feasible enough to parse Korean comparative sentences and yield proper semantic representations though further development is needed for a finer model for contextual information.

• The Journal of the Acoustical Society of Korea
• /
• v.25 no.7
• /
• pp.312-318
• /
• 2006

The conventional lip-synch system has a two-step process, speech segmentation and recognition. However, the difficulty of speech segmentation procedure and the inaccuracy of training data set due to the segmentation lead to a significant Performance degradation in the system. To cope with that, the connected vowel recognition method using Head-Body-Tail (HBT) model is proposed. The HBT model which is appropriate for handling relatively small sized vocabulary tasks reflects co-articulation effect efficiently. Moreover the 7 vowels are merged into 3 classes having similar lip shape while the system is optimized by employing a class dependent SCHMM structure. Additionally in both end sides of each word which has large variations, 8 components Gaussian mixture model is directly used to improve the ability of representation. Though the proposed method reveals similar performance with respect to the CHMM based on the HBT structure. the number of parameters is reduced by 33.92%. This reduction makes it a computationally efficient method enabling real time operation.

• Journal of the Institute of Convergence Signal Processing
• /
• v.2 no.4
• /
• pp.9-15
• /
• 2001

In this paper we investigate context dependent acoustic models to improve the performance of the Korean speech recognition . The algorithm are using the Korean phonological rules and decision tree, By Successive State Splitting(SSS) algorithm the Hidden Merkov Netwwork(HM-Net) which is an efficient representation of phoneme-context-dependent HMMs, can be generated automatically SSS is powerful technique to design topologies of tied-state HMMs but it doesn't treat unknown contexts in the training phoneme contexts environment adequately In addition it has some problem in the procedure of the contextual domain. In this paper we adopt a new state-clustering algorithm of SSS, called Phonetic Decision Tree-based SSS (PDT-SSS) which includes contexts splits based on the Korean phonological rules. This method combines advantages of both the decision tree clustering and SSS, and can generated highly accurate HM-Net that can express any contexts To verify the effectiveness of the adopted methods. the experiments are carried out using KLE 452 word database and YNU 200 sentence database. Through the Korean phoneme word and sentence recognition experiments. we proved that the new state-clustering algorithm produce better phoneme, word and continuous speech recognition accuracy than the conventional HMMs.