• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.177-180
• /
• 2000

기존의 제한적 단어 인식과는 달리 무제한 단어 음성인식에 있어서는 방대한 용량의 단어 모델을 참조로 인식이 이루어지게 되어, 참조모델과 입력패턴과의 비교를 위한 탐색시간이 너무 길어지게 된다. 본 논문에서 제한하는 방법은 무제한 단어 음성인식 시스템을 구축하기 위해 선행되어야 하는 모음열 사전을 구축하는 것이다. 음성인식시 입력패턴과 참조모델에 속한 모든 단어와의 비교를 수행하지 않고, 입력패턴의 모음열을 인식한 후, 인식된 모음열 단어들만을 참조모델에서 인식 후보로 두어 인식을 수행하게 하여 시간적인 측면에서의 효율성을 기하는 것이다. 결과적으로 본 연구 방법은 무제한 단어 음성인식에서의 실시간 처리라는 점에 주 목적을 두었다.

• Proceedings of the Acoustical Society of Korea Conference
• /
• spring
• /
• pp.103-106
• /
• 2000

무제한 인식 시스템을 구현하기 위해서는 적절한 인식단위, 훈련 데이터 베이스의 확보, 인식단위의 분할, 인식 알고리즘과 같은 문제점을 모두 해결하여야 한다. 따라서 본 논문에서는 무제한 음성인식 시스템의 인식의 기본 단위로 모음의 안정구간을 검출하여 분할하는 CV(Consonant-Vowel), VC(Vowel-Consonant), VC CV(Vowel-Consonant-Consonant-Vowel)단위와 분할 파라미터를 제안하고, 분할 실험을 통해 그 유효성을 확인하고자 한다.

• Proceedings of the Korea Institute of Convergence Signal Processing
• /
• 2000.08a
• /
• pp.181-184
• /
• 2000

본 논문에서는 영한 음차 변환을 이용한 음성인식 및 합성기를 구현하였다. 음성인식의 경우 CV(Consonant Vowel), VCCV, VCV, VV, VC 단위를 사용하였다. 위의 단위별로 미리 구축된 모델을 결합함으로써 무제한 음성인식 시스템을 구축하였다. 따라서 영한 음차 변환을 이용하게 되면 인식 대상이 영어단어일 경우에도 이를 한글 발음으로 변환한 후 그에 해당하는 모델을 생성함으로써 인식이 가능하다. 음성 합성기의 경우 합성에 필요한 한국어 음성 데이터 베이스를 구축하고, 입력되는 텍스트에 따라 이를 연결하여 합성음을 생성한다. 영어가 입력될 경우 영한 음차 변환을 이용하여 입력된 영어발음을 한글로 바꾸어 준 후 입력하게 되므로 별도의 영어 합성기 없이도 합성음을 생성할 수 있다.

• Journal of Animal Science and Technology
• /
• v.48 no.6
• /
• pp.871-880
• /
• 2006

This study was conducted to estimate the maximum amount and time of milk consumption by Holstein calves during two months of pre- weaning period and to compare ad libitum milk feeding with conventional system for their effects on dry feed consumption, nutrient intake, growth performance and occurrence of diarrhea during the pre-weaning phase. Holstein calves (n=20) were alternatively fed colostrum at 10% of their body weight or ad libitum for the first three days (d) and then assigned either to conventional (n=10; calves were fed milk at 10% of their body weight from birth to weaning at d 60 of their age) or ad libitum milk feeding (n=10; calves were fed ad libitum milk from birth to weaning at d 60 of their age). Milk intake as a percent of body weight was increased in ad libitum fed calves from 13.59% at d 1 to 23.50% at d 30 of their age, thereafter it was gradually reduced. Ad libitum fed calves consumed daily 9.8kg milk compared with 4.3kg milk in restricted fed calves(p<0.05). Daily starter and mixed grass hay intake was gradually increased in both groups with the age however; the group differences were noted non-significant up to 40 d of age. Overall from birth to weaning conventionally fed calves ate 18.4, 20 and 18.9% more dry matter, crude protein and total digestible nutrients, respectively than ad libitum milk fed calves (p<0.05). Weight gain was 49, 18.4 and 26% higher in ad libitum milk fed calves than conventionally fed calves during the first month, the second month and from birth to weaning, respectively(p<0.05). Body weight was not significantly different between groups for the first 20 d of age. However, it was significantly higher for ad libitum milk fed calves at d 30, 40, 50, 60 (weaning) and 90 (post weaning) of their age(p<0.05). Similar body weights for both conventionally and ad libitum milk fed calves were observed for the post weaning period at 120, 150 and 180 d of age. Feed efficiency was significantly higher for conventionally fed calves than ad libitum milk fed calves(p<0.05). No sign of diarrhea, fever or cough was observed in calves fed either conventionally or ad libitum milk during the experimental period. In conclusion, calves fed more milk remained healthy and gained higher weight before weaning and maintained this advantage at least up to 90 d of their age.

• Proceedings of the KSPS conference
• /
• 2003.10a
• /
• pp.175-179
• /
• 2003

In order to produce a high quality (intelligibility and naturalness) synthesized speech, it is very important to get an accurate grapheme-to-phoneme conversion and prosody model. In this paper, we analyzed Chinese texts using a segmentation, POS tagging and unknown word recognition. We present a grapheme-to-phoneme conversion using a dictionary-based and rule-based method. We constructed a prosody model using a probabilistic method and a decision tree-based error correction method. According to the result from the above analysis, we can successfully select and concatenate exact synthesis unit of syllables from the Chinese Synthesis DB.

• Journal of the Korean Society of Costume
• /
• v.55 no.2 s.92
• /
• pp.70-81
• /
• 2005

This paper researches the system of official uniforms and characteristics of the system based on Ming-Shi(明史), Da-Ming-Hui-Dian(大明會典) and Hong-Wu-Li-Zhi(洪武禮制) in the emperor Hong-wu(洪武帝) of Ming dynasty(明). The system could be divided in to three terms, i.e., the early (initial), the middle and the end (last) terms. In the first(1368) & the third year of the emperor's reign (1370) the system of official uniforms was simple f9r his people not to be luxurious and for revived the system of official uniforms of the Han dynasty(漢) and annulled the system of the Yuan dynasty(元), be-cause the country was not stable yet. During the middle term of Hong-wu, the 14th (1381) to 16th (1383) year of his reign, people attained luxurious lives and developed a strong sense of rank as the country became more stable. As a result, the system became more complicated when the emperor used the costumes to represent political and social ranking. In the end of the emperor's age, the 23rd (1390) to 26th (1393), it reveals his strong intention to establish a system of costumes that separated the ruler from the ruled. He wanted to strengthen the power of Emperor for his little grandson, Jian-wen(建文帝), who would succeed to the throne and would have absolute power. These changes became fundamental to the Ming dynasty's costume system and the Ming dynasty kept them for long time.