Earticle

The Chinese dash, pozhehao (破折号，⎯⎯), serves ten different functions, including providing supplementary information, shifting topics, summarizing, and concluding. By contrast, the Korean dash (—) fulfills only one function according to the current Korean Orthography (2017). In Lu Xun’s novels, pozhehao performs various functions and is often considered to add aesthetic effects that transcend the text itself. This highlights Lu Xun’s meticulous use of pozhehao to effectively convey his authorial intentions. In the complete collection of Lu Xun’s novels, pozhehao appears 372 times. In two Korean translations of this collection, it was rendered as the Korean dash 318 times and 145 times, respectively, revealing a significant disparity. This suggests that translating pozhehao as the Korean dash may lead to divergent interpretations between the two versions. It is therefore essential to conduct comprehensive research on the translation of pozhehao and its functions. This study aims to outline the functions of pozhehao based on current punctuation rules and to propose effective methods for translating it into Korean.

In this study, we carried out a series of experiments to explore how ChatGPT (version 4o) evaluated Korean-English translations. Using two datasets of human translations (n=57) and two datasets of post-edited translations (n=56), all drawn from Lee and Lee (2021), we adopted two evaluation approaches with strict prompt control. In Experiment A, ChatGPT rated the four datasets freely on a five-point scale without specific criteria. In Experiment B, which was conducted concurrently with Experiment A, ChatGPT rated the same datasets using a prescribed, criterion-referenced five-point scale. To assess intra-rater reliability, we repeated both experiments one month later. This study yielded both quantitative and qualitative findings, including the following: (1) ChatGPT’s average scores differed significantly from those of human raters; (2) correlations between human and ChatGPT scores ranged from ‘moderate’ to ‘strong’; (3) the use of the prescribed rating scale improved ChatGPT’s reliability as a rater; (4) ChatGPT exhibited very low intra-rater reliability; and (5) ChatGPT’s self-justifications for its ratings varied in quality, often failing to identify obvious errors.

This study examines the rapidly expanding field of audiovisual translation, with a particular focus on Subtitling for d/Deaf and Hard of Hearing (SDH) audiences, a service increasingly in demand as media accessibility receives greater societal emphasis. The recent surge in over-the-top platforms and the popularity of Korean content have driven demand for interlingual SDH, though it has traditionally been approached as an intralingual endeavor. To better understand interlingual SDH, this study compares it with regular subtitles and contrasts the practices between intralingual and interlingual SDH. The analysis focuses on Netflix's Hellbound Season 2, which integrated feedback from a focus group of d/Deaf and Hard of Hearing individuals during production, as well as its publicly available guidelines. Findings reveal that while both subtitles and SDH operate under spatial and temporal constraints, they differ in their approach to relevance and redundancy. Additionally, approaches to editing and reduction vary between intralingual and interlingual SDH. Effective interlingual SDH creation demands not only translation skills but also a catered understanding of the target viewers' needs. This study aims to provide practical support for translators and seeks to benefit SDH viewers by enhancing the effectiveness of subtitles for accessibility.

This study focuses on the development of a learning management s ystem (LMS)-based automatic annotation authoring tool for computer -assisted interpreter training (CAIT). The primary goal of the research is to establish a specialized LMS that integrates AI-powered speech-to -text (STT) and automatic annotation functionalities to create an effici ent feedback system. Through the use of external APIs (Application P rogramming Interface), the tool implements automatic detection of di sfluency elements such as fillers, pauses, and repetitions in STT-gen erated interpreting outputs, which are visualized using a TextAE (Tex t Annotation Editor) program. Additionally, the system enhances obj ectivity in evaluation by providing visualized assessment results and supports autonomous practice for learners through a self-study featu re. The significance of this study lies in its effective application of AI technologies to interpreter training, enabling an integrated system th at encompasses task creation, execution, and evaluation in one platf orm. It is anticipated that this research will contribute to the digital tr ansition of interpreter training and improve educational efficiency.

This study investigated the potential benefits and limitations of automated interpreting assessment by reviewing various approaches, including machine translation (MT) quality metrics and quality estimation (QE), followed by an experimental application of MTQE to a small-scale dataset of English-Korean consecutive interpretations by three interpreting students. CometKiwi, one of the state-of-the-art MTQE methods that show strong evaluation performance without requiring reference translations, was employed to compare automated evaluation with human evaluation. The findings reveal that automated evaluations exhibited a strong correlation with human evaluations and achieved complete agreement in ranking interpreting outputs, confirming the feasibility of QE-based automated evaluation for interpretations. At both the text and segment levels, higher-quality interpreting outputs showed greater alignment between human and automated evaluations, while lower-quality outputs tended to receive relatively higher scores from the QE model, highlighting discrepancies with human evaluations. It was noted that CometKiwi returned scores above zero for uninterpreted segments, possibly overestimating the final overall quality scores. The study suggests that automated evaluation scores could serve as useful resources for interpreting educators and students when an independent reference source is needed to support human evaluation.

This study investigates user perceptions and experiences of machine translation (MT) through text mining analyses of user reviews, conceptualized as digital paratexts of translation, for Google Translate, Papago, and DeepL. Employing keyword extraction, sentiment analysis, and selective qualitative analysis on 4,913 reviews, this study identifies key factors influencing user engagement and technology acceptance, including translation quality, usability, and social influences. The findings indicate that users perceive Google Translate as beneficial for its extensive language coverage, Papago as effective for language learning due to its user-friendly tools, and DeepL as superior in accuracy and naturalness. Despite generally positive attitudes toward MT, users highlight critical areas for improvement, such as interface usability and limited language support. Moreover, the reviews reflect broader socio-cultural dynamics, illustrating how societal narratives shape MT adoption. This study underscores the complementary role of MT alongside human translation and offers practical insights for developers, practitioners, educators, and researchers. By exploring user-driven insights, this research advances understanding of the evolving landscape of MT and its integration into professional and educational contexts.

Pivot subtitling, a form of audiovisual translation, involves translating the source text into the target text using an intermediary language. This study examines how pivot subtitling—translating through an intermediary language—affects the Filipino subtitles of the Korean series Squid Game when English serves as the pivot language. It analyzes how using an English pivot text influences the transfer of linguistic and cultural nuances across languages. The qualitative analysis in this research focuses on three categories of data, namely politeness markers, address terms, and cultural references, all rich with socio-pragmatic meanings. The findings reveal significant omissions and distortions of the original Korean cultural nuances in the Filipino target text, compromising the depth of character relationships and cultural resonance. This may affect the Filipino audience's interpretation of the story and characters and their emotional engagement. This research underscores the limitations of pivot subtitling in multilingual contexts by highlighting specific examples. It also emphasizes the urgent need for improvements in the translation process to enhance linguistic accuracy and cultural fidelity, filling in research gaps in translating Korean content into Filipino and applying pivot subtitling in this context.

This study compares how Papago, Google Translate, and ChatGPT perform when translating police dialogues, using BLEU scores, manual assessment, and error analysis. The analysis results are as follows. All three tools showed low BLEU scores compared to reference translations. In manual evaluation, ChatGPT significantly outperformed others, scoring 8.6 out of 10, compared to Google Translate's 6.5 and Papago's 5.6. Error analysis confirmed ChatGPT's superiority, with only 41 errors, while Google Translate and Papago produced 123 and 152 errors respectively. Across all tools, substitution was the most common accuracy error, followed by omission and addition. These results suggest that ChatGPT is the most reliable tool for police communication with foreign nationals.

Pansori, a UNESCO-recognized Korean musical performance, combines narration (aniri), song (chang), and gestures (ballim). This study examines how Pansori’s chang can be translated into English while maintaining its rhythms and melodies. Focusing on Mother, a creative English Pansori composed by master singer So-ra Kim, performed at the 48th Pansori Festival (New York, 2018), this study uses text analysis and interviews. The analysis explores how English lyrics are composed to match Pansori’s rhythms; in jungmori, most important content words are matched with the first (hap) and ninth beats (cheok); in jungjungmori, content words are placed on the first (deong) and seventh beats (gung); and in eotmori, content words match the first (deong) and sixth beats (gung), with the rhythm extended to 20 beats. Finally, this study emphasizes the crucial role of translators in making Pansori accessible to global audiences.