• 한국소성가공학회:학술대회논문집
• /
• 한국소성가공학회 2003년도 The Korea-Japan Plastics Processing Joint Seminar
• /
• pp.1-6
• /
• 2003

In injection molding of an optical disk, a toric lens, etc., their performance depends on the transcription preciseness of fine surface structure of a mold. However, transcription behavior has not been made clear yet, because transcription is made in very short time and the structure is very small. In this paper, transcription properties have been examined, by using V-grooves of various sizes. machined on mold surfaces, and the following results are obtained. (1) Transcription properties have been made clear experimentally and it was found that the mold temperature $T_D$ makes great influence on the transcription property and that compression applying time $t_c$ should be taken more than 2.0s for fine transcription. (2) A mechanical model of transcription process, in consideration with strain recovery due to viscoelastic property of polymer. is proposed. (3) Simulation results agree with experimental ones fairly well. It means that the transcription model is useful for estimation of transcription property in advance of an actual. injection molding.

• 음성과학
• /
• 제10권4호
• /
• pp.23-40
• /
• 2003

This paper examines transcription of sounds from a phonological perspective. It has found that most of transcriptions have been done on a segmental basis alone, without consideration of the whole phonological systems and levels, and without a full understanding of the nature of the linguistic and phonetic alphabets. In a word, sound transcriptions have not been done on the basis of the phonology of the language and the alphabet. This study shows a phonological model for transcribing foreign and native sounds, suggesting ways of improving some of the current transcription systems such as the Hangeul transcription of loan words and the romanization of Hangeul, as well as the phonetic transcription of English and other foreign languages.

• 음성과학
• /
• 제9권1호
• /
• pp.111-135
• /
• 2002

An annotation model of a non-native speech database has been devised, wherein English is the target language and Korean is the native language. The proposed annotation model features overt transcription of predictable linguistic information in native speech by the dictionary entry and several predefined types of error specification found in native language transfer. The proposed model is, in that sense, different from other previously explored annotation models in the literature, most of which are based on native speech. The validity of the newly proposed model is revealed in its consistent annotation of 1) salient linguistic features of English, 2) contrastive linguistic features of English and Korean, 3) actual errors reported in the literature, and 4) the newly collected data in this study. The annotation method in this model adopts the widely accepted conventions, Speech Assessment Methods Phonetic Alphabet (SAMPA) and the TOnes and Break Indices (ToBI). In the proposed annotation model, SAMPA is exclusively employed for segmental transcription and ToBI for prosodic transcription. The annotation of non-native speech is used to assess speaking ability for English as Foreign Language (EFL) learners.

• 대한음성학회지:말소리
• /
• 제61호
• /
• pp.87-100
• /
• 2007

A punctuation generation system which combines prosodic information with acoustic and language model information is presented. Experiments have been conducted first for the reference text transcriptions. In these experiments, prosodic information was shown to be more useful than language model information. When these information sources are combined, an F-measure of up to 0.7830 was obtained for adding punctuation to a reference transcription. This method of punctuation generation can also be applied to the 1-best output of a speech recogniser. The 1-best output is first time aligned. Based on the time alignment information, prosodic features are generated. As in the approach applied in the punctuation generation for reference transcriptions, the best sequence of punctuation marks for this 1-best output is found using the prosodic feature model and an language model trained on texts which contain punctuation marks.

• The Journal of the Acoustical Society of Korea
• /
• 제21권1E호
• /
• pp.3-11
• /
• 2002

Broadcast news transcription is one of the hardest tasks in speech recognition because broadcast speech signals have much variability in speech quality, channel and background conditions. We developed a Korean broadcast news speech recognizer. We used a morpheme-based dictionary and a language model to reduce the out-of·vocabulary (OOV) rate. We concatenated the original morpheme pairs of short length or high frequency in order to reduce insertion and deletion errors due to short morphemes. We used a lexicon with multiple pronunciations to reflect inter-morpheme pronunciation variations without severe modification of the search tree. By using the merged morpheme as recognition units, we achieved the OOV rate of 1.7% comparable to European languages with 64k vocabulary. We implemented a hidden Markov model-based recognizer with vocal tract length normalization and online speaker adaptation by maximum likelihood linear regression. Experimental results showed that the recognizer yielded 21.8% morpheme error rate for anchor speech and 31.6% for mostly noisy reporter speech.

• 한국음향학회:학술대회논문집
• /
• 한국음향학회 2002년도 하계학술발표대회 논문집 제21권 1호
• /
• pp.87-90
• /
• 2002

This paper analyzes statistically the relationship between size and balance of text corpus by evaluation of the effect of interview sentences in language model for Korean broadcast news transcription system. Our Korean broadcast news transcription system's ultimate purpose is to recognize not interview speech, but the anchor's and reporter's speech in broadcast news show. But the gathered text corpus for constructing language model consists of interview sentences a portion of the whole, $15\%$ approximately. The characteristic of interview sentence is different from the anchor's and the reporter's in one thing or another. Therefore it disturbs the anchor and reporter oriented language modeling. In this paper, we evaluate the effect of interview sentences in language model for Korean broadcast news transcription system and analyze statistically the relationship between size and balance of text corpus by making an experiment as the same procedure according to varying the size of corpus.

• 한국음향학회지
• /
• 제43권2호
• /
• pp.207-213
• /
• 2024

본 논문은 기계학습 기반 자동악보전사 모델의 입력에 음악적인 정보를 추가하는 방법을 통해 원하는 성능 향상을 얻는 방법을 다루었다. 여기서, 추가한 음악적인 정보는 각 시간 단위마다 발생하는 음고 개수 정보이며, 이는 정답지에서 활성화되는 음고 개수를 세는 방법으로 획득한다. 획득한 음고 개수 정보는 기존 모델의 입력인 로그 멜-스펙트로그램 아래에 연결하여 사용했다. 본 연구에서는 네 가지 음악 정보를 예측하는 네 종류의 블록이 포함된 자동악보전사 모델을 사용하였으며, 각 블록이 예측해야하는 음악 정보에 해당하는 음고 개수 정보를 기존의 입력에 추가해주는 간단한 방법이 모델의 학습에 도움이 됨을 확인했다. 성능 개선을 검증하기 위하여 MIDI Aligned Piano Sounds(MAPS) 데이터를 활용하여 실험을 진행하였으며, 그 결과 모든 음고 개수 정보를 활용할 경우 프레임 기준 F1 점수에서 9.7 %, 끝점을 포함한 노트 기준 F1 점수에서 21.8 %의 성능 향상을 확인하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제26권4호
• /
• pp.813-826
• /
• 2015

본 연구는 화학 반응 모형의 추정 문제를 다루고 있다. 화학 반응 모형이란 생화학 분야에서 종(species) 들 간의 상호작용을 통한 변화 과정을 설명하기 위한 모형으로 생화학 분야 뿐 만 아니라 질병의 확산과정을 설명하는데 적용하는 모형이다. 본 연구에서는 화학 반응 모형 안에서 종들의 움직임이 확률적이라는 가정하에 Gillespie 알고리즘을 이용하여 모형 추정을 위한 우도함수를 구축하였다. 제한적인 자료구조 하에서 베이지안 접근법에 기반하여 MCMC (Markov chain Monte Carlo)방법에 기반한 모수의 추정 방법을 제안하였다. 제안된 방법들은 생태계 포식자-피식자 관계를 설명하기 위한 Lotka-Volterra 모형과 유전자 전사 (gene transcription) 과정을 설명하기 위한 L1 retrotransposition 모형에 적용하였다. 그 결과 우수한 추정 결과를 보였다.

• Asian Pacific Journal of Cancer Prevention
• /
• 제16권2호
• /
• pp.437-442
• /
• 2015

Aim: To establish a pancreatic cancer stem cell model using human pancreatic cancer cells in nude mice to provide a platform for pancreatic cancer stem cell research. Materials and Methods: To establish pancreatic cancer xenografts using human pancreatic cancer cell line SW1990, nude mice were randomly divided into control and gemcitabine groups. When the tumor grew to a volume of $125mm^3$, they treated with gemcitabine at a dose of 50mg/kg by intraperitoneal injection of 0.2ml in the gemcitabine group, while the mice in control group were treated with the same volume of normal saline. Gemcitabine was given 2 times a week for 3 times. When the model was established, the proliferation of pancreatic cancer stem cells was observed by clone formation assay, and the protein and/or mRNA expression of pancreatic stem cell surface markers including CD24, CD44, CD133, ALDH, transcription factors containing Oct-4, Sox-2, Nanog and Gli, the key nuclear transcription factor in Sonic Hedgehog signaling pathway was detected by Western blot and/or RT-PCR to verify the reliability of this model. Results: This model is feasible and safe. During the establishment, no mice died and the weight of nude mice maintained above 16.5g. The clone forming ability in gemcitabine group was stronger than that of the control group (p<0.01). In gemcitabine group, the protein expression of pancreatic cancer stem cell surface markers including CD44, and ALDH was up-regulated, the protein and mRNA expression of nuclear transcription factor including Oct-4, Sox-2 and Nanog was also significantly increased (P<0.01). In addition, the protein expression of key nuclear transcription factor in Sonic Hedgehog signaling pathway, Gli-1, was significantly enhanced (p<0.01). Conclusions: The pancreatic cancer stem cell model was successfully established using human pancreatic cancer cell line SW1990 in nude mice. Gemcitabine could enrich pancreatic cancer stem cells, simultaneously accompanied by the activation of Sonic Hedgehog signaling pathway.

• 한국음향학회지
• /
• 제42권2호
• /
• pp.102-111
• /
• 2023

자동 음악 채보는 주어진 오디오에서 음표 정보를 추출하는 태스크로, 이 연구에서는 피아노 음악의 자동음악 채보 모델에서 지연 시간을 줄이는 방법을 소개한다. 신경망 기반 채보 모델이 피아노 채보에도 적용되어 높은 정확도를 기록하였고 이를 이용한 실시간 구현도 소개된 바 있지만, 채보를 위한 지연 시간이 길어 인터랙티브 시나리오에서 활용하기에 한계가 있었다. 이 문제를 해결하기 위해 본 연구는 Fast Fourier Transformation(FFT)에서 윈도우 크기와 홉 크기를 줄이거나 합성곱 레이어의 커널 크기를 수정하고 시간 축에서 레이블을 이동하여 모델이 시작을 더 일찍 예측하도록 훈련하는 등 피아노 전사를 위한 신경망의 내재적 지연 시간을 줄이는 몇 가지 기술을 제안한다. 실험 결과, 이러한 접근 방식을 결합하면 높은 전사 정확도를 유지하면서 지연 시간을 줄일 수 있음을 알 수 있었다. 기존 모델은 160 ms의 지연 시간을 가지고 음표 F1 점수는 93.43 %였으나 제안한 방법을 적용하면 96 ms와 64 ms의 지연 시간 동안 각각 92.67 %와 90.51 %의 F1 점수를 달성할 수 있었다. 이러한 결과는 향후 피아노 교육을 위한 실시간 피드백 제공 등 다양한 인터랙티브 시나리오를 위한 자동 채보 모델에 활용될 수 있을 것이다.