• The Journal of the Acoustical Society of Korea
• /
• v.13 no.1
• /
• pp.24-31
• /
• 1994

In this paper, we report on the development of a speaker independent continuous speech recognition system using continuous hidden Markov models. The continuous hidden Markov model consists of mean and covariance matrices and directly models speech signal parameters, therefore does not have quantization error. Filter bank coefficients with their 1st and 2nd-order derivatives are used as feature vectors to represent the dynamic features of speech signal. We use the segmental K-means algorithm as a training algorithm and triphone as a recognition unit to alleviate performance degradation due to coarticulation problems critical in continuous speech recognition. Also, we use the one-pass search algorithm that Is advantageous in speeding-up the recognition time. Experimental results show that the system attains the recognition accuracy of $83\%$ without grammar and $94\%$ with finite state networks in speaker-indepdent speech recognition.

• Journal of Korea Multimedia Society
• /
• v.4 no.6
• /
• pp.524-533
• /
• 2001

Efficient continuous speech recognition system for practical applications requires that the processing be carried out in real time and high recognition accuracy. In this paper, we study the acoustic models by adopting the PDT-SSS algorithm and the language models by iterative learning so as to improve the speech recognition accuracy. And the adaptive pruning algorithm is applied to the continuous speech. To verify the effectiveness of proposed method, we carried out the continuous speech recognition for the Korean air flight reservation task. Experimental results show that the adopted algorithm has the average 90.9% for continuous speech recognition and the average 90.7% for word recognition accuracy including continuous speech. And in case of adopting the adaptive pruning algorithm to continuous speech, it reduces the recognition time of about 1.2 seconds(15%) without any loss of accuracy. From the result, we proved the effectiveness of the PDT-SSS algorithm and the adaptive pruning algorithm.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.21 no.4
• /
• pp.525-537
• /
• 2020

This study aimed to determine the current state and characteristics of simulation-based operating processes in nursing education based on the Jeffries theoretical framework in South Korea by taking an integrated look at study findings in order to provide a scientific basis for future simulation-based operating processes. We searched eight databases, including the Korea Education and Research Information Service, National Library, Korean Studies Information Service System, National Digital Science Library, Korea Institute of Science and Technology Information, KOREAMED, and Korean Medical Database, using terms "simulation" and "nursing" as keywords in November 2017 in the Korean language. Sixteen studies were identified, reviewed, and appraised in this integrative review. The literature was categorized into these themes: general study characteristics, operation method, teaching and learning methods, subject characteristics, outcome variables, and theoretical framework. The simulation processes in nursing education in South Korea that were analyzed in this study did not fully reflect the main concepts suggested in the NLN Jeffries simulation framework. Thus, simulation program developers need to consider and incorporate a variety of strategies, based on the identification of essential components, to improve simulation effectiveness.

• The Journal of the Acoustical Society of Korea
• /
• v.31 no.4
• /
• pp.235-243
• /
• 2012

This paper studies characteristics and causes of sound, and many others by distinguishing passivity and activity of the cases of a dog vocalizing human's words. As a result of the previous cases of vocalization of human's words, the dog was able to understand characteristics of a host's voice and imitate the sound using his own vocal organs. This is the case of passive vocalization accompanied by temporary voice imitation without a function of communication. On the contrary, as a consequence of the recently reported case in which a dog vocalizes such words as "Um-ma" and "Nu-na-ya," it shows the vocalization pattern clearly distinguished from the prior cases. The given dog repeatedly vocalizes pertaining words in an active manner according to circumstances and plays a role of fundamental communication and interaction with its host. The reason why the dog can vocalize the man's words actively is determined to be that the dog has a high level of intelligence and intimacy with its host, that people react actively to its pertaining pronunciation, and so forth. The following results can be used for the study that investigates animals' sound with vocalization possibility and language learning feasibility.

• Proceedings of the Acoustical Society of Korea Conference
• /
• autumn
• /
• pp.29-32
• /
• 2000

본 논문에서는 음성인식 시스템의 음향모델 개선을 위한 기초적 연구로서, 문맥적인 요소를 필요로 하는 SSS(Successive State Splitting)와 필요로 하지 않는 SSS-free 알고리즘을 이용한 HMnet(Hidden Markov Network) 음향모델 작성방법에 대해 검토하고 작성한 음향모델을 한국어에 적용하여 그 유효성을 확인하였다. HMnet을 이용한 음소모델의 작성방법은 전체 학습 데이터에 대해서 각각 2개의 상태를 가지는 초기 모델을 작성한 후, 이를 시간과 문맥방향으로의 최대 분포를 가지는 상태를 재분할한 후 임의의 상태수가 될 때까지 상태분할을 계속적으로 수행케 하여 각 음소모델을 작성하게 된다. 작성한 HMnet 음향모델의 유효성을 확인하기 위해 ETRI 445 단어의 3인에 대한 화자종속 음소인식 실험을 수행하였다. 인식실험 결과, SSS 알고리즘을 이용한 화자종속실험의 경우 상태수 520에서 평균 $62.8\%$의 인식률을, SSS-free 알고리즘의 경우 상태수 420에서 평균 $64.2\%$의 인식률을 얻었다. 이 결과는 HMM을 이용한 경우(약$43.4\%$)보다 $20\%$이상의 인식률 향상을 보여 이 알고리즘의 유효성을 확인할 수 있었다. SSS와 SSS-free를 비교한 경우, SSS-free가 SSS보다 낮은 상태수에서 평균 $1.4\% 향상된 인식률을 보였다.

• Annual Conference on Human and Language Technology
• /
• 2006.10e
• /
• pp.268-272
• /
• 2006

질의응답 시스템은 사용자 질의에 해당하는 정답을 찾기 위해서 세부 분류된 개체명을 사용한다. 이러한 세부 분류 개체명 인식을 위해서 대부분의 시스템이 일반 대분류 개체명인식 후에 사전 등을 이용하여 세부 분류로 나누는 방법을 이용하고 있다. 본 논문에서는 질의응답 시스템을 위한 세부 분류 개체명 인식을 위해서 Conditional Random Fields를 이용한다. 개체명 인식의 과정을 개체명 경계 인식과 경계가 인식된 개체명의 클래스 분류의 두 단계로 나누어, 개체명 경계 인식에 Conditional Random Fields를 이용하고, 경계 인식된 개체명의 클래스 분류에는 Maximum Entropy를 이용한다. 실험결과 147개의 세부분류 개체명 인식에 대해서 정확도 85.8%, 재현률 81.1%. F1=83.4의 성능을 얻었고. baseline model 보다 학습 시간이 27%로 줄고 성능은 증가하였다. 또한 제안된 세부 분류개체명 인식기를 이용하여 질의응답 시스템에 적용한 결과 26%의 성능향상을 보였다.

• Journal of Korean Institute of Industrial Engineers
• /
• v.40 no.1
• /
• pp.18-33
• /
• 2014

A Word is the smallest unit for text analysis, and the premise behind most text-mining algorithms is that the words in given documents can be perfectly recognized. However, the newly coined words, spelling and spacing errors, and domain adaptation problems make it difficult to recognize words correctly. To make matters worse, obtaining a sufficient amount of training data that can be used in any situation is not only unrealistic but also inefficient. Therefore, an automatical word extraction method which does not require a training process is desperately needed. WordRank, the most widely used unsupervised word extraction algorithm for Chinese and Japanese, shows a poor word extraction performance in Korean due to different language structures. In this paper, we first discuss why WordRank has a poor performance in Korean, and propose a customized WordRank algorithm for Korean, named KR-WordRank, by considering its linguistic characteristics and by improving the robustness to noise in text documents. Experiment results show that the performance of KR-WordRank is significantly better than that of the original WordRank in Korean. In addition, it is found that not only can our proposed algorithm extract proper words but also identify candidate keywords for an effective document summarization.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.3
• /
• pp.273-278
• /
• 2004

In this paper, the recurrent neural networks (RNN) is applied to compensate for HMM recognition algorithm, which is commonly used as main recognizer. Among these recurrent neural networks, the multi-layer recurrent neural prediction model (MRNPM), which allows operating in real-time, is used to implement learning and recognition, and HMM and MRNPM are used to design a hybrid-type main recognizer. After testing the designed speech recognition algorithm with Korean number pronunciations (13 words), which are hardly distinct, for its speech-independent recognition ratio, about 5% improvement was obtained comparing with existing HMM recognizers. Based on this result, only optimal (recognition) codes were extracted in the actual DSP (TMS320C6711) environment, and the embedded speech recognition system was implemented. Similarly, the implementation result of the embedded system showed more improved recognition system implementation than existing solid HMM recognition systems.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.11
• /
• pp.2125-2130
• /
• 2016

In this paper, we researched the excellency of 3D graphic nursery contents which is developed by mobile AR technology. AR technology has currently people's attention because of the potential to be core contents of future ICT industry. We applied AR nursery contents for kid's subtitle language selection in Korean, Chinese and English education. The original fairy tale consisted of 6~8 scenes for the 3D contents production, and was adapted and translated. Dubbing was dubbed by the native speaker using the standard pronunciation, and the effect sound was edited separately to fit the scene. After composing a scenario, constructing a 3D model, constructing a interaction, constructing a sound effect, and creating content metadata, the Unity 3D game engine is executed to create a project and describe it as a script. It provides a fun and informative tradition of fairy tales with abundant content that incorporates ICT technology, accepting advanced technology-based education, and having opportunities to perceive software in daily life.

• Journal of KIISE
• /
• v.43 no.7
• /
• pp.773-780
• /
• 2016

Semantic information and features are very important for Semantic Role Labeling(SRL) though many SRL systems based on machine learning mainly adopt lexical and syntactic features. Previous SRL research based on semantic information is very few because using semantic information is very restricted. We proposed the SRL system which adopts semantic information, such as named entity, word sense disambiguation, filtering adjunct role based on sense, synonym cluster, frame extension based on synonym dictionary and joint rule of syntactic-semantic information, and modified verb-specific numbered roles, etc. According to our experimentations, the proposed present method outperforms those of lexical-syntactic based research works by about 3.77 (Korean Propbank) to 8.05 (Exobrain Corpus) F1-scores.