• Journal of the Korea Computer Industry Society
• /
• v.3 no.5
• /
• pp.585-594
• /
• 2002

This paper, describes about creation and implementation of phonetic value- transcription conversion table for postprocessing of the voice recognition. Transcription set generator, which produces transcription set that is pronounced as recognized phonetic value, is designed and implemented to postprocess for the voice recognition system which recognizes syllable unit phonetic value Phonetic value-transcription conversion table is produced with transcription-phonetic value conversion table produced by modeling standard pronunciation on petrinet. To show that phonetic value-transcription conversion table produces correct transcription set, transcription set generator is designed and implemented. This paper proves that correct transcription set is produced, which is including pre-vocalization transcription as a result of experimenting standard pronunciation examples and the words randomly sampled from pronunciation dictionary.

• Phonetics and Speech Sciences
• /
• v.9 no.3
• /
• pp.7-15
• /
• 2017

The purpose of this study was to investigate the different aspects of word-initial liquid sounds in Korean according to generations. Five women in their 50s and seven in their 20s participated in the experiment. We examined FL (formant of liquids) and voice sustained time by using Praat software. Three English native speakers were asked to judge the Korean speakers' recorded speech samples for marking [l] or [r] using evaluation sheet. The results of the two experiments revealed three important aspects. First, there was a statistically significant difference between the two groups in the FL of the words 'racket' and 'ruby.' Second, we found statistically significant differences in 'rhythm', 'ruby' and 'litter' from the measurement of the duration of the acoustic data. Third, there was no difference in pronunciation between the two groups according to the phonemes of the original language. The results of this study showed that it is difficult to say that the duration of word-initial liquids and the phoneme difference of the original language are indicators to distinguish the word-initial liquids between generations. Also, it was seen that the pronunciation of Korean word-initial liquid sounds varied across generations.

• KIPS Transactions on Software and Data Engineering
• /
• v.1 no.2
• /
• pp.91-102
• /
• 2012

Computer aided language learning has become popular. But the level of automation of constructing a Korean learning assistant system is not so high because a practical language learning system needs large scale knowledge resources, which is very hard to acquire. In this paper, we propose a Korean learning assistant system that utilizes easily obtainable knowledge resources like a corpus, web documents and a lexicon. Our system has three modules - problem solving, pronunciation marker and writing assistant. Automatic problem generator uses a corpus and a lexicon to make problems with one correct answer and three distracters, then verifies their suitability by utilizing frequency information from web documents. We analyze pronunciation rules for a pronunciation marker and recommend appropriate words and sentences in real-time by using data extracted from a corpus. In experiment, we evaluate 400 automatically generated problems, which show 89.9% problem suitability and 64.9% example suitability.

• Journal of the Korea Computer Industry Society
• /
• v.3 no.12
• /
• pp.1679-1690
• /
• 2002

This paper, describe about a generation of a syllable restoration regulation for a post processing of a speech recognition and a decrease of a restoration candidate. It created a syllable restoration regulation to create a restoration candidate pronounced with phonetic value recognized through a post processing of the formula system that was a tone to recognize syllable unit phonetic value for a performance enhancement of a dialogue serial speech recognition. Also, I presented a plan to remove a regulation to create unused notation from a real life in a restoration regulation with a plan to reduce number candidate of a restoration meeting. A design implemented a restoration candidate set generator in order a syllable restoration regulation display that it created a proper restoration candidate set. The proper notation meeting that as a result of having proved about a standard pronunciation example and a word extracted from a pronunciation dictionary at random, the notation that an utterance was former was included in proved with what a generation became.

• The KIPS Transactions:PartB
• /
• v.11B no.1
• /
• pp.99-106
• /
• 2004

The syllable restoration is a process restoring a phonetic value recognized in a speech recognition device with the notation form that a vocalization is former. In this paper a syllable restoration rule was composed of a based on standard pronunciation for a syllable restoration process. A syllable restoring regulation was used, and a generation method of a notation candidate set was researched. Also, A study is held to reduce the number of created notation candidate. Three phases of reduction processes were suggested. Reduction of a notation candidate has the non-notation syllable, non-vocabulary syllable and non-stem syllable. As a result of experiment, an average of 74％ notation candidate decrease rates were shown.

• The KIPS Transactions:PartB
• /
• v.9B no.6
• /
• pp.827-834
• /
• 2002

In this paper, we proposed the formation of Korean-Roman alphabet notation conversion table for the generation of Korean-Roman alphabet notation that also meets revised Roman alphabet notation. Introduced a mathematical analyzing method of the natural language which used a petrinet model so that a base of Roman alphabet notation analyzed standard pronunciation and Roman alphabet notation to work mathematically. It display the practical example through a petrinet modeling of a plan and Roman alphabet notation to create a Korean Roman alphabet notation conversion table with the method of the analysis that used a petrinet model, and present a mathematical modeling plan and application method of Korean. We developed application program based on window in order to verify a created Korean-Roman alphabet notation conversion table, and compared the result of an application program with Roman alphabet notation of an Roman alphabet notation example dictionary.

• Journal of Korea Multimedia Society
• /
• v.18 no.8
• /
• pp.943-949
• /
• 2015

A reader usually spends a lot of time browsing and searching word meaning in a dictionary, internet or smart applications in order to find the unknown words. In this paper, we propose a method to compensate this drawback. The proposed method introduces a vocabulary upon recognizing a word or group of words that was captured by a smart phone camera. Through this proposed method, organizing and editing words that were captured by smart phone, searching the dictionary data using bisection method, listening pronunciation with the use of speech synthesizer, building and editing of vocabulary stored in database are given as the features. A smart phone application for organizing English words was established. The proposed method significantly reduces the organizing time for unknown English words and increases the English learning efficiency.

• Proceedings of the Korean Society for Cognitive Science Conference
• /
• 2005.05a
• /
• pp.233-238
• /
• 2005

본 논문에서는 학습을 이용한 발음 변이 모델링을 통해 특정 영역에 최적화된 발음 사전 자동 생성의 방법을 제시하였다. 학습 방법을 이용한 발음 변이 모델링의 오류를 최소화 하기 위하여 본 논문에서는 발음 변이 규칙의 적응 기법을 도입하였다. 발음 변이 규칙의 적응은 대용량 음성 말뭉치에서 발음 변이 규칙을 유도한 후, 상대적으로 작은 용량의 음성 말뭉치에서 유도한 규칙과의 결합을 통해 이루어 진다. 본 논문에서 사용된 발음 사전은 해당 형태소의 앞 뒤 음소 문맥의 음운 현상을 반영한 발음 사전이며, 학습 방법으로 얻어진 발음 변이 규칙을 대용량 문자 말뭉치에 적용하여 해당 형태소의 발음을 자동 생성하였다. 발음 사전의 평균 발음의 수는 적용된 발음 변이 규칙의 확률 값들의 한계 값 조정에 의해 이루어졌다. 기존의 지식 기반의 발음 사전과 비교 할 때, 본 방법론으로 작성된 발음 사전을 이용한 대화체 음성 인식 실험에서 0.8%의 단어 오류율(WER)이 감소하였다. 또한 사전에 포함된 형태소의 평균 발음 변이 수에서도 기존의 방법론에서 보다 5.6% 적은 수에서 최상의 성능을 보였다.

• Proceedings of the KIEE Conference
• /
• 1990.07a
• /
• pp.47-50
• /
• 1990

Deafs cannot speak straight voice as normal people in lack of feedback of their pronunciation, therefore speech training is required. In this study, fundamental frequency, intensity, formant frequencies, vocal tract graphic and vocal tract area function, extracted from speech signal, are used as feature parameter. AR model, whose coefficients are extracted using inverse filtering. is used as speech generation model. In connect ion between vocal tract graphic and speech parameter, articulation distances and articulation distance functions in selected 15-intervals are determined by extracted vocal tract areas and formant frequencies.

• Annual Conference on Human and Language Technology
• /
• 2007.10a
• /
• pp.210-215
• /
• 2007

본 연구에서는 코퍼스 확률에 기반하여 한국어 표준 발음 생성에 대한 연구를 한다. 기존의 이은영 외 (2005)에서 연구된 규칙기반의 한국어 IPA 발음 변환방식과는 달리 본 연구에서는 음운변환 코퍼스를 바탕으로 표준발음을 변환한다. 이 방식을 위해서 Brill(1995)에서 제안한 변형기반 학습방식이 활용되었으며, 단계적인 처리방식이 아닌 입-출력 대응 방식의 확률적 처리 방식이 제안되었다. 음운변환 방식은 음운규칙에 근거한 처리가 아닌 언어자원인 코퍼스를 활용해서 처리하였다는 점에서 기존의 연구방식과 차이가 있다. 또한, 기존 연구에서는 음운규칙을 단계적으로 적용하여서 입력형이 출력형으로 도출되기 위해서 여러 단계를 거쳤지만, 본 연구에서는 입력형과 출력형의 일대일 대응이라는 점에서 차이점을 보인다.