• Annual Conference of KIPS
• /
• 2024.05a
• /
• pp.466-467
• /
• 2024

본 논문은 전문 스킬 트레이너들의 개인 농구 기술 분석 및 피드백 영상에 더빙을 자동으로 적용하는 시스템을 제안한다. 이 시스템은 농구 용어집 기반 번역, 음성-텍스트 변환 모델 간의 비교 분석, 영상과 더빙 트랙 동기화 알고리즘을 통해 다양한 언어로의 신속한 자동 번역과 더빙을 가능하게 함으로써 선수와 코치 간의 언어 장벽 없는 소통을 지원한다. 본 연구는 자동 더빙 기술에 힘입어 원격 농구 교육 효율성과 질의 재고 및 저변 확산에 기여하고자 한다.

• Journal of Digital Convergence
• /
• v.16 no.12
• /
• pp.459-466
• /
• 2018

Gendered innovations is a term used by policy makers and academics to refer the process of creating better research and development (R&D) for both men and women. In this paper, we analyze the literatures in image and speech signal processing that can be used in ICT, examine the importance of gendered innovations through case study. Therefore the latest domestic and foreign literature related to image and speech signal processing based on gender research is searched and a total of 9 papers are selected. In terms of gender analysis, research subjects, research environment, and research design are examined separately. Especially, through the case analysis of algorithms of the elderly voice signal processing, machine learning, machine translation technology, and facial gender recognition technology, we found that there is gender bias in existing algorithms, and which leads to gender analysis is required. We also propose a gendered innovations method integrating sex and gender analysis in algorithm development. Gendered innovations in ICT can contribute to the creation of new markets by developing products and services that reflect the needs of both men and women.

• Journal of Korea Society of Industrial Information Systems
• /
• v.9 no.3
• /
• pp.64-70
• /
• 2004

Korean and Japanese have same structure of sentences because they belong to same family of languages. So, The transfer driven machine translation is most efficient to translate each other. This paper introduce a method which creates a transfer dictionary for Token Based Transfer Driven Koran-Japanese Machine Translation(TB-TDMT). If the transfer dictionaries are created well, we get rid of useless effort for traditional parsing by performing shallow parsing. The semi-parser makes the dependency tree which has minimum information needed output generating module. We constructed the transfer dictionaries by using the corpus obtained from ETRI spoken language database. Our system was tested with 900 utterances which are collected from travel planning domain. The success-ratio of our system is $92\%$ on restricted testing environment and $81\%$ on unrestricted testing environment.

• Journal of the Korean Society of Radiology
• /
• v.10 no.1
• /
• pp.59-64
• /
• 2016

The purpose of this study intend to develop the automatic voice system for language communicating over foreign patients in testing of magnetic resonance imaging. Used sentences for the automatic voice system was selected with twelve sentences and five foreign languages in hospital in use. These translated texts were provided free of charge from Oddcast's website, which is recorded each language-specific voice as Wav files, The recorded audio file were produced by the two types of a power-point show of MS-office 2013(extension; ppsx) form and Web-enabled system using the PHP program to be applied in the mobile phone environment. It is considered to relieve the stress about an language barrier to medical technician and the voice system designed with the variable language will be enhanced the diagnostic information on foreign patients. Because automatic voice system designed for this study can enhance the understanding of test in between korean medical technician and foreign patients, It will be expected to utilize more usefully in clinical practice.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.178-185
• /
• 2021

Sequence to sequence(S2S) 기반 음성인식 후처리기를 훈련하기 위한 학습 데이터 구축을 위해 (음성인식 결과(speech recognition sentence), 전사자(phonetic transcriptor)가 수정한 문장(Human post edit sentence))의 병렬 말뭉치가 필요하며 이를 위해 많은 노동력(human-labor)이 소요된다. BackTranScription (BTS)이란 기존 S2S기반 음성인식 후처리기의 한계점을 완화하기 위해 제안된 데이터 구축 방법론이며 Text-To-Speech(TTS)와 Speech-To-Text(STT) 기술을 결합하여 pseudo 병렬 말뭉치를 생성하는 기술을 의미한다. 해당 방법론은 전사자의 역할을 없애고 방대한 양의 학습 데이터를 자동으로 생성할 수 있기에 데이터 구축에 있어서 시간과 비용을 단축 할 수 있다. 본 논문은 BTS를 바탕으로 제주어 도메인에 특화된 음성인식 후처리기의 성능을 향상시키기 위하여 모델 수정(model modification)을 통해 성능을 향상시키는 모델 중심 접근(model-centric) 방법론과 모델 수정 없이 데이터의 양과 질을 고려하여 성능을 향상시키는 데이터 중심 접근(data-centric) 방법론에 대한 비교 분석을 진행하였다. 실험결과 모델 교정없이 데이터 중심 접근 방법론을 적용하는 것이 성능 향상에 더 도움이 됨을 알 수 있었으며 모델 중심 접근 방법론의 부정적 측면 (negative result)에 대해서 분석을 진행하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2002.04b
• /
• pp.451-453
• /
• 2002

연어는 습관적으로 같이 자주 나타나는 단어열로 각 단어로 분리하기보다 통합해 처리하는 것이 효율적이기 때문에 기계 번역과 음성 인식등에서 유용만 정보로 사용된다. 이러한 연어를 추출하기 위해 본 논문에서는 2가지 경우를 고려했는데, 첫 번째로 인어를 말뭉치에 자주 나타나는 단어열이라고 했을 때 단어열들의 엔트로피가 일정값 이상이면 연어로 추출했다 두 번째로 통사적 제약이 있는 연어를 주술하기 위해 앞 또는 뒤에 올 단어를 제약하는 단어의 엔트로피를 구해 일정값 미만이면 그 단어를 포함한 단어열을 연어로 추출했다. 실험은 품사 부착된 HANTCE 말뭉치를 가지고 수행했고, 젓 번째 방법으로 실험했을 때 엔드로피가 2이상인 단어열을 가지고 분리된 연어도 유도해냈다.

• Proceedings of the KSLP Conference
• /
• 1993.12a
• /
• pp.13-13
• /
• 1993

실어증 감별진단검사법은 실어증환자들의 진단, 치료계획의 설정, 치료 효과 및 예후의 판정 등을 위하여 사용된다. 그러나 아직 우리나라에는 우리에게 알맞은 실어증 감별진단 검사법이 개발되지 않았으므로 본 연구에서는 정신문화 및 언어구조면에서 우리나라와 유사성이 가장 높다고 생각되는 일본에서 널리 사용되고 있는 실어증 감별진단검사법(노인총합 연구소판)을 우리나라의 정신문화 및 언어상의 특성을 고려하여 번역 보완환 다음 한국어판 실어증 감별진단검사법을 시안(I)을 실어증 환자군과 정상 대조군에 적용하여 그 가치를 평가하고자 하였다. (중략)

• MALSORI
• /
• no.61
• /
• pp.31-48
• /
• 2007

This study investigated the roles of name agreement and word frequency in the English-Korean word translation task. Using the low-frequency homonyms with low name agreement as stimuli, Experiment 1 revealed that the name agreement of materials is a determinant which could modulate times to translate English words into Korean equivalents. On the contrary, Experiment 2 showed that the name agreement of materials does not play a decisive role in the translation task, using the low-frequency homonyms having high name agreement as stimuli. In Experiment 3, we identified that the frequency effects observed from previous two experiments are indeed brought about during the lexical access. Our findings suggest that the word frequencies of materials have a strong influence on English-Korean word translation times, and homonyms are represented independently each other in the lexeme level.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2016.10a
• /
• pp.644-646
• /
• 2016

Deaf are it is difficult to communicate to represent the voice heard, so theay use mostly using the speech, sign language, writing, etc. to communicate. It is the best way to use sign language, in order to communicate deaf and normal people each other. But they must understand to use sign language. In this paper, we designed and implementated finger language translation system to support communicate between deaf and normal people. We used leap motion as input device that can track finger and hand gesture. We used raspberry pi that is low power sing board computer to process input data and translate finger language. We implemented application used Node.js and MongoDB. The client application complied with HTML5 so that can be support any smart device with web browser.

• Annual Conference of KIPS
• /
• 2019.10a
• /
• pp.552-554
• /
• 2019

Deaf mutism (a hearing-impaired person and speech disorders) communicates using sign language. There are difficulties in communicating by voice. However, sign language can only be limited in communicating with people who know sign language because everyone doesn't use sign language when they communicate. In this paper, a finger language translation system is proposed and implemented as a means for the disabled and the non-disabled to communicate without difficulty. The proposed algorithm recognizes the finger language data by leap motion and self-learns the data using machine learning technology to increase recognition rate. We show performance improvement from the simulation results.