• Journal of the Korea Computer Industry Society
• /
• v.2 no.6
• /
• pp.867-874
• /
• 2001

In this paper, considering the pronunciation characteristic of Korean, recognition units which can shorten the recognition time and reflect the coarticulation effect simultaneously are suggested. These units are composed of monophone and hipbone ones. Monophone units are applied to the vowels which represent stable characteristic. Biphones are used to the consonant which vary according to adjacent vowel. In the experiment of word recognition of PBW445 database, the compound units result in comparable recognition accuracy with 57％ speed up compared with triphone units and better recognition accuracy with similar speed. In addition, we can reduce the memory size because of fewer units.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.273-275
• /
• 2016

외국인 유학생과 국내 체류 외국인을 포함하여 한국어를 학습하고자 하는 외국인이 지속적으로 증가함에 따라, 외국인 한국어 학습자의 교육에 대한 관심도 높아지고 있다. 기존 맞춤법 검사기는 한국어를 충분히 이해할 수 있는 한국인의 사용에 중점을 두고 있어, 외국인 한국어 학습자가 사용하기에는 다소 부적절하다. 본 논문에서는 한국어의 문맥 특성과 외국인의 작문 특성을 반영한 한국어 교정 방식을 제안한다. 제안하는 시스템에서는 말뭉치에서 추출한 어절 바이그램에 대한 음절 역색인을 구성하여 추천 표현을 빠르게 제시할 수 있으며, 키보드 후킹에 기반한 사용자인터페이스를 제공하여 사용자 편의를 높인다.

• The Journal of the Acoustical Society of Korea
• /
• v.13 no.6
• /
• pp.5-14
• /
• 1994

This paper is about an experiment of speaker-independent automatic Korean spoken words recognition using Multi-Layered Perceptron and Error Back-propagation algorithm. The object words are 50 citynames of D.D.D local numbers. 43 of those are 2 syllables and the rest 7 are 3 syllables. The words were not segmented into syllables or phonemes, and some feature components extracted from the words in equal gap were applied to the neural network. That led independent result on the speech duration, and the PARCOR coefficients calculated from the frames using linear predictive analysis were employed as feature components. This paper tried to find out the optimum conditions through 4 differerent experiments which are comparison between total and pre-classified training, dependency of recognition rate on the number of frames and PAROCR order, recognition change due to the number of neurons in the hidden layer, and the comparison of the output pattern composition method of output neurons. As a result, the recognition rate of $89.6\%$ is obtaimed through the research.

• KIISE Transactions on Computing Practices
• /
• v.22 no.3
• /
• pp.119-126
• /
• 2016

In this study, we verified the feasibility of a Korean morphological analyzer that uses a pre-analyzed Eojeol dictionary and syllable-based probabilistic model. For the verification, MACH and KLT2000, Korean morphological analyzers, were cloned with a pre-analyzed eojeol dictionary and syllable-based probabilistic model. The analysis results were compared between the cloned morphological analyzer, MACH, and KLT2000. The 10 million Eojeol Sejong corpus was segmented into 10 sets for cross-validation. The 10-fold cross-validated precision and recall for cloned MACH and KLT2000 were 97.16%, 98.31% and 96.80%, 99.03%, respectively. Analysis speed of a cloned MACH was 308,000 Eojeols per second, and the speed of a cloned KLT2000 was 436,000 Eojeols per second. The experimental results indicated that a Korean morphological analyzer that uses a pre-analyzed eojeol dictionary and syllable-based probabilistic model could be used in practical applications.

• Journal of KIISE:Software and Applications
• /
• v.34 no.2
• /
• pp.169-177
• /
• 2007

Hangul text normalization in current Unicode makes wrong Hangul syllable problems when using with precomposed modern Hangul syllables and composing old Hangul by using conjoining-Hangul Jamo and compatibility Hangul Jamo. This problem comes from allowing incorrect normalization form of compatibility Hangul Jamo and Hangul Symbol and also permitting to use conjoining-Hangul Jamo mixture with precomposed Hangul syllable in Unicode Hangul composing rule. It is caused by lack of consideration of old Hangul and/or insufficient understanding of Hangul code processing when writing specification for normalization forms in Unicode. Therefore on this paper, we study Hangul code in Unicode environment, specifically problems of normalization used for Web and XML, IDN in nowadays. Also we propose modification of Hangul normalization methods and Hangul composing rules for correct processing of Hangul normalization in Unicode.

• The Journal of the Acoustical Society of Korea
• /
• v.23 no.2
• /
• pp.162-172
• /
• 2004

This study investigates the following three things through a production experiment and acoustic analysis: 1) relationship between speech rate and the segment duration in Korean, 2) relationship between speech rate and spectral characteristics of vowels, i. e. undershoot, and 3) correlation between the vowel duration and undershoot. The results showed that the faster the speech rate nab, the shorter the duration of syllables and segments was. A few speakers were affected by speech rate in the durational ratios between closure and aspiration in a stop and between Towel and consonant in a syllable. Closure duration and vowel duration were more affected compared to aspiration and consonant duration, respectively. Speakers showed some differences in the extent to which speech rate affected vowel undershoot, implying that speakers used different production mechanisms for spectral characteristics of vowels: Some speakers speeded up movement of articulatory organs according to speech rate increase while some kept it constant regardless of speech rate change.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.10
• /
• pp.165-172
• /
• 2010

There are two method of error correction in vocabulary recognition system. one error pattern matting base on method other vocabulary mean pattern base on method. They are a failure while semantic of key-word problem for error correction. In improving, in this paper is propose system of key-word error correction using algorithm of syllable restoration. System of key-word error correction by processing of semantic parse through recognized phoneme meaning. It's performed restore by algorithm of syllable restoration phoneme apply fluctuation before word. It's definitely parse of key-word and reduced of unrecognized. Find out error correction rate using phoneme likelihood and confidence for system parse. When vocabulary recognition perform error correction for error proved vocabulary. system performance comparison as a result of recognition improve represent 2.3% by method using error pattern learning and error pattern matting, vocabulary mean pattern base on method.

• Journal of Korea Multimedia Society
• /
• v.9 no.1
• /
• pp.23-32
• /
• 2006

This paper describes the optimization of a language model and an acoustic model to improve speech recognition using Korean unit digits. Since the model is composed of a finite state network (FSN) with a disyllable, recognition errors of the language model were reduced by analyzing the grammatical features of Korean unit digits. Acoustic models utilize a demisyllable pair to decrease recognition errors caused by inaccurate division of a phone or monosyllable due to short pronunciation time and articulation. We have used the K-means clustering algorithm with the transformed successive state splitting in the feature level for the efficient modelling of feature of the recognition unit. As a result of experiments, 10.5% recognition rate is raised in the case of the proposed language model. The demi-syllable fair with an acoustic model increased 12.5% recognition rate and 1.5% recognition rate is improved in transformed successive state splitting.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2003.11a
• /
• pp.408-411
• /
• 2003

정보 검색에서 우리가 자주 쓰는 단어는 표준어인 경우도 있고 통신어인 경우도 있고 외국어인 경우도 있다. 그러나 표준어가 아닌 다른 언어로 검색을 하면 다른 결과가 나타날 수 있다. 예를 들어 컴퓨터에 관한 정보를 찾을 때 ‘컴퓨타’로 검색을 하면 다른 검색 결과가 나오게 된다. 우리나라에서 현재 쓰이고 있는 말들은 이런 애매한 발음의 외래어가 많이 생성되고 소멸된다. 그러므로 이런 외래어들을 전부 사전에 등록할 수는 없고 설사 등록한다 하더라도 용량과 검색시간만 늘어나게 된다. 본 논문에서는 검색엔진에서 이런 외래어에 대한 인식 성능을 높이기 위해 외래어 사전 없이 외래어를 인식하는 방안을 제시한다.

• Annual Conference on Human and Language Technology
• /
• 2010.10a
• /
• pp.9-14
• /
• 2010

본 논문에서는 명사의 띄어쓰기 bigram과 단일명사 정보를 이용하여 복합명사를 분해하는 방법을 제시한다. 붙여쓰기와 띄어쓰기를 모두 허용하는 복합명사의 특징에 따라 띄어쓰기 bigram으로 후보를 선정할 경우, 분해시간과 후보의 수를 크게 줄일 수 있으며, 긴 음절의 복합명사도 bigram의 chain을 통해 빠르게 후보 조합이 가능하다. 분해 후보가 복수일 경우, 명사 간 bigram 확률을 계산하여 최적의 분해 후보를 선정한다.