• ETRI Journal
• /
• v.43 no.5
• /
• pp.775-786
• /
• 2021

To translate in real time, a simultaneous translation system should determine when to stop reading source tokens and generate target tokens corresponding to a partial source sentence read up to that point. However, conventional attention-based neural machine translation (NMT) models cannot produce translations with adequate latency in online scenarios because they wait until a source sentence is completed to compute alignment between the source and target tokens. To address this issue, we propose a reinforced learning (RL)-based attention mechanism, the reinforced attention mechanism, which allows a neural translation model to jointly train the stopping criterion and a partial translation model. The proposed attention mechanism comprises two modules, one to ensure translation quality and the other to address latency. Different from previous RL-based simultaneous translation systems, which learn the stopping criterion from a fixed NMT model, the modules can be trained jointly with a novel reward function. In our experiments, the proposed model has better translation quality and comparable latency compared to previous models.

• ETRI Journal
• /
• v.45 no.1
• /
• pp.18-27
• /
• 2023

We present an English-Korean speech translation corpus, named EnKoST-C. End-to-end model training for speech translation tasks often suffers from a lack of parallel data, such as speech data in the source language and equivalent text data in the target language. Most available public speech translation corpora were developed for European languages, and there is currently no public corpus for English-Korean end-to-end speech translation. Thus, we created an EnKoST-C centered on TED Talks. In this process, we enhance the sentence alignment approach using the subtitle time information and bilingual sentence embedding information. As a result, we built a 559-h English-Korean speech translation corpus. The proposed sentence alignment approach showed excellent performance of 0.96 f-measure score. We also show the baseline performance of an English-Korean speech translation model trained with EnKoST-C. The EnKoST-C is freely available on a Korean government open data hub site.

• ETRI Journal
• /
• v.37 no.3
• /
• pp.541-550
• /
• 2015

In this paper, we propose a classification-based approach for hybridizing statistical machine translation and rulebased machine translation. Both the training dataset used in the learning of our proposed classifier and our feature extraction method affect the hybridization quality. To create one such training dataset, a previous approach used auto-evaluation metrics to determine from a set of component machine translation (MT) systems which gave the more accurate translation (by a comparative method). Once this had been determined, the most accurate translation was then labelled in such a way so as to indicate the MT system from which it came. In this previous approach, when the metric evaluation scores were low, there existed a high level of uncertainty as to which of the component MT systems was actually producing the better translation. To relax such uncertainty or error in classification, we propose an alternative approach to such labeling; that is, a cut-off method. In our experiments, using the aforementioned cut-off method in our proposed classifier, we managed to achieve a translation accuracy of 81.5% - a 5.0% improvement over existing methods.

• (The)Study of the Eastern Classic
• /
• no.71
• /
• pp.259-304
• /
• 2018

In this paper, we analyzed and analyzed the entry path and the capacity of translation competent personnel who participated in the translation project for the last 5 years. For this purpose, the entry path of the translation workforce is classified as 'employee', 'research course', 'vt', and 'external expert'. And the number of manpower and the amount of translated manuscripts were examined by year. In addition, the average amount of manpower for one translator was also examined. As a result of the research and analysis, the proportion of the total translation companies in the translation business was in the order of external experts - from the research course - from the qualification examination - from the staff. The proportion of manuscripts is in the order of from the research course - external expert - from the qualification examination - from the staff. In addition, we surveyed and analyzed the present status of each of the translators and discussed the characteristics and problems of each business sector. In the translation of historical documents, the proportion of manpower is the largest in the research process, but the proportion of manuscripts is the highest in the qualification test, and that of the experts in the translation and special classics is significant. In addition, it was judged that translation work of translator was proceeding inefficiently by analyzing average manpower quantity of translation worker. In order to solve these problems, three plans were proposed: 'Need to plan the use of translation personnel based on business plan', 'Improvement of qualification examination for cross-country and translation committee members', 'Training of translation experts in specialized fields and expansion of entry routes'. Each of these measures must be validated and validated through detailed analysis and research. However, it is hoped that the translator will help the translation project by establishing a more systematic and planned business plan and a plan to utilize the translation workforce.

• BMB Reports
• /
• v.50 no.4
• /
• pp.186-193
• /
• 2017

In mammals, cap-dependent translation of mRNAs is initiated by two distinct mechanisms: cap-binding complex (CBC; a heterodimer of CBP80 and 20)-dependent translation (CT) and eIF4E-dependent translation (ET). Both translation initiation mechanisms share common features in driving cap- dependent translation; nevertheless, they can be distinguished from each other based on their molecular features and biological roles. CT is largely associated with mRNA surveillance such as nonsense-mediated mRNA decay (NMD), whereas ET is predominantly involved in the bulk of protein synthesis. However, several recent studies have demonstrated that CT and ET have similar roles in protein synthesis and mRNA surveillance. In a subset of mRNAs, CT preferentially drives the cap-dependent translation, as ET does, and ET is responsible for mRNA surveillance, as CT does. In this review, we summarize and compare the molecular features of CT and ET with a focus on the emerging roles of CT in translation.

• Proceedings of the Korean Society for Language and Information Conference
• /
• 2007.11a
• /
• pp.105-114
• /
• 2007

This paper addresses a method for customizing an English-to-Korean machine translation system from general domain to patent domain. The customizing method consists of following steps: 1) linguistically studying about characteristics of patent documents, 2) extracting unknown words from large patent documents and constructing large bilingual terminology, 3) extracting and constructing the patent-specific translation patterns 4) customizing the translation engine modules of the existing general MT system according to linguistic study about characteristics of patent documents, and 5) evaluating the accuracy of translation modules and the translation quality. This research was performed under the auspices of the MIC (Ministry of Information and Communication) of Korean government during 2005-2006. The translation accuracy of the customized English-Korean patent translation system is 82.43% on the average in 5 patent fields (machinery, electronics, chemistry, medicine and computer) according to the evaluation of 7 professional human translators. In 2006, the patent MT system started an on-line patent MT service in IPAC (International Patent Assistance Center) under MOCIE (Ministry of Commerce, Industry and Energy) in Korea. In 2007, KIPO (Korean Intellectual Property Office) tries to launch an English-Korean patent MT service.

• Physical Therapy Korea
• /
• v.25 no.1
• /
• pp.53-61
• /
• 2018

Background: The process of language translation, adaptation, and cross-cultural validation of tools for use in multiple countries requires the adoption of well-established, comprehensive, and rigorous methodological approaches. Back translation, which is the most recommended method, permits the detection of errors in the translation and the identification of words or phrases that cannot be accurately or literally translated. Objects: The aim of this study was to verify the content validity of a Korean version of a Balance Evaluation Systems test (BESTest) by using a back-translation method. Methods: This research was conducted in six steps: 1) translation of the BESTest into Korean, 2) evaluation of the translation conformity of Korean-translated BESTest, 3) evaluation of the degree of translation comprehension, 4) back translation of Korean BESTest, 5) evaluation of the technical and conceptual equivalence, and 6) completion of the Korean version of BESTest by the translation verification committee. Results: In this study, Korean version of the BESTest achieved a rating of more than 3 (moderate) for translation comprehension, and technical equivalence and conceptual equivalence of back translation were evaluated as 3 (moderate) or more. Conclusion: The Korean version of the BESTest has proven content validity and is an appropriate tool to measure balance function.

• Physical Therapy Korea
• /
• v.22 no.4
• /
• pp.51-61
• /
• 2015

The purpose of this study were to translate the Fullerton Advanced Balance (FAB) scale into Korean and to verify the content validity by utilizing a back-translation method with a view to assessing balance function and the risk of falling in a clinical research setting. This research was conducted in six steps. First, three Korean physical therapists translated the FAB scale into Korean. Second, two bilingual professors of physical therapy and a physical therapist evaluated translation conformity of Korean-translated FAB scale. In the third and fourth steps, twelve physical therapists evaluated the degree of translation comprehension, and a translator back-translated the Korean FAB scale into the original language. Fifth, a bilingual professor of physical therapy and two native speakers evaluated the technical and conceptual equivalence between the original and translation versions. In this process, inappropriate translated items were revised using recommended substitute words or sentences, and all items were evaluated on the basis of three points or more on a rating scale in terms of translation comprehension, and the technical and conceptual equivalence of the back-translation. In the sixth and last step, the translation verification committee completed the final Korean version. The above process indicated that the content validity of the Korean-translated FAB scale was established by means of systematic translation methods, and it can therefore be used to assess balance function and the risk of falls in a clinical research setting.

• Language and Information
• /
• v.18 no.1
• /
• pp.77-97
• /
• 2014

Multilingual machine translation means the machine translation that is for more than two languages. Common transfer means the transfer in which we can reuse the transfer rules among similar languages according to linguistic typology. Therefore, the multilingual machine translation based on common transfer is the multilingual machine translation that can share the transfer rules among languages with similar linguistic typology. This paper describes the linguistic modeling for multilingual machine translation based on common transfer under development. This linguistic modeling consists of the linguistic devices such as 1) multilingual common Part-of-Speech set, 2) multilingual common transfer format, 3) multilingual common transfer chunking, and 4) multilingual common transfer rules based on linguistic typology. Validity of this linguistic modeling for multilingual machine translation is shown in the simulation. The multilingual machine translation system based on common transfer including Korean, English, Chinese, Spanish, and French will be developed till 2018.

• Language and Information
• /
• v.2 no.2
• /
• pp.185-220
• /
• 1998

The purpose of this research is to propose a set of basic guidelines for the construction of English test suites, a set of basic guidelines for the construction of Korean test suites to objectively evaluate the performance of machine translation systems. For this end, we constructed 650 English test sentences, 650 Korean test sentences, and developed the statistical methods and tools for the comparative evaluation of the English-Korean machine translation systems. It also evaluates the existing commercial English-Korean machine translation systems. The importance of this research lies in that it will promote an awareness of the importance and need of testing machine translation systems within the Natural Language Community. This research will also make a big contribution to the development of evaluation methods and techniques for appropriate test suites for Korean information processing systems. The results of this research can be used by the natural language community to test the performance and development of their information processing systems or machine translation systems.