Search results

Mr Patrick in his paper has called for a liberal policy on the part of companies in making translations available; J shall now describe one method of combining liberality with enlightened self‐interest. Since I believe the BISITS to be the first successful one of its kind, I shall describe how it was formed and developed as well as how it works today.

Discusses the vital role of technical translation services in the process of information dissemination and technology transfer with particular reference to India. Argues that, for the most effective provision of translation services, subject specialists should be available in special libraries and documentation centres. Gives suggestions for improving translation services in India.

Multilingual user guides, manuals and thesauri, and more linguistically universal classification schemes need to be produced to help overcome the language barrier when searching for references. Such developments however would be of only limited value unless the full document can be made available in a language known to the user. In some countries centres have been founded to provide help with the identification of existing translations, these centres include the International Translation Centre. Published translations include translated books reports and journals while unpublished translations are those which are completed on an ad hoc basis. Bibliographic control is achieved by listing in abstracting/indexing journals or on bibliographic data bases. Publications including Journals in Translation for translated journals and World Transindex for ad hoc translations, can be used to trace the existence of a translation. The availability of the translated document itself varies according to the two types of translations. Published translations are reasonably accessible while ad hoc translations are scattered over many libraries and translating agencies. Libraries can help at both the stage of acquiring the reference and identifying and obtaining the documents.

This paper discusses the use of raw machine translations in an IT research and development environment. Researchers use machine translation as a drafting tool for scientific papers. The language pairs are German→ English and English→ German. The raw machine translations are produced on an experimental basis by means of the MT systems LOGOS, METAL and Globalink Power Translator Professional. The experiments are to show whether the systems are suitable for this purpose or not. Since the use of raw machine translations is considered to be crucial to the future of MT, the paper discusses success or failure of the technology against this background.

The Internet has the potential to facilitate understanding across cultures and languages by removing the physical barriers to intercultural communication. One possible contributor to this development has been the recent release of freely‐available automated direct machine translation systems, such as AltaVista with SYSTRAN, which translates from English to five other European languages (French, German, Italian, Spanish and Portuguese), and vice versa. However, concerns have recently been raised over the performance of these systems, and the potential for confusion that can be created when the intended meaning of sentences is not correctly translated (i.e. semantic processing errors). In this paper, we use an iterative paradigm to examine errors associated with interlingual divergence in meaning arising from the automated machine translation of English proverbs. The need for the development of Web‐based translation systems, which have an explicit cross‐linguistic representation of meaning for successful intercultural communication, is discussed.

In his book The foreign language barrier in sciences and technology (Aslib, 1962), C. W. Hanson wrote about ‘the desirability of using one international language for scientific publication, whether it be an existing natural language or a constructed one’. ‘This global approach to the language problem has some support,’ he continues, ‘but even if international agreement were achieved progress would be slow and it would probably be several decades before the international language was in anything approaching universal use.’ Exactly ten years later, Prof. A. I. Mikhailov, director of the All Union Institute for Scientific and Technical Information (VINITI) concludes, speaking about ways to overcome linguistic barriers: ‘the final solution of the problem lies in developing a universal language of science which will develop into a single language common to the world scientific community’ (Invited Papers of the 36th International Congress of FID, Budapest, 1972). After ten years the problem of overcoming linguistic barriers in international exchange of information by means of introduction of a universal language appears to be as vivid as a full decade ago.

This article examines the role of BLDSC's Translations Section as a focal point in the national and international availability of translations. BLDSC's collection of translations into English is amongst the largest in the world (the Translations Index has over half a million entries) and the Centre is the principal source both of translations and of information about them in the UK. Translations are acquired from all over the world, covering most languages, with the main source languages emerging as Russian, German, French and Japanese. On average, 40,000 requests are received by the Translations Section each year, the majority treating BLDSC as the first and only resort. 11% of these requests are from overseas. The Centre also publishes several major bibliographical tools in this field, and has close links with other national and international translations centres. It acts not only as a document supplier but also as an information provider in this specialized area.

The paper aims to enhance Arabic machine translation (MT) by proposing novel approaches: (1) a dimensionality reduction technique for word embeddings tailored for Arabic text, optimizing efficiency while retaining semantic information; (2) a comprehensive comparison of meta-embedding techniques to improve translation quality; and (3) a method leveraging self-attention and Gated CNNs to capture token dependencies, including temporal and hierarchical features within sentences, and interactions between different embedding types. These approaches collectively aim to enhance translation quality by combining different embedding schemes and leveraging advanced modeling techniques.

Recent works on MT in general and Arabic MT in particular often pick one type of word embedding model. In this paper, we present a novel approach to enhance Arabic MT by addressing three key aspects. Firstly, we propose a new dimensionality reduction technique for word embeddings, specifically tailored for Arabic text. This technique optimizes the efficiency of embeddings while retaining their semantic information. Secondly, we conduct an extensive comparison of different meta-embedding techniques, exploring the combination of static and contextual embeddings. Through this analysis, we identify the most effective approach to improve translation quality. Lastly, we introduce a novel method that leverages self-attention and Gated convolutional neural networks (CNNs) to capture token dependencies, including temporal and hierarchical features within sentences, as well as interactions between different types of embeddings. Our experimental results demonstrate the effectiveness of our proposed approach in significantly enhancing Arabic MT performance. It outperforms baseline models with a BLEU score increase of 2 points and achieves superior results compared to state-of-the-art approaches, with an average improvement of 4.6 points across all evaluation metrics.

The proposed approaches significantly enhance Arabic MT performance. The dimensionality reduction technique improves the efficiency of word embeddings while preserving semantic information. Comprehensive comparison identifies effective meta-embedding techniques, with the contextualized dynamic meta-embeddings (CDME) model showcasing competitive results. Integration of Gated CNNs with the transformer model surpasses baseline performance, leveraging both architectures' strengths. Overall, these findings demonstrate substantial improvements in translation quality, with a BLEU score increase of 2 points and an average improvement of 4.6 points across all evaluation metrics, outperforming state-of-the-art approaches.

The paper’s originality lies in its departure from simply fine-tuning the transformer model for a specific task. Instead, it introduces modifications to the internal architecture of the transformer, integrating Gated CNNs to enhance translation performance. This departure from traditional fine-tuning approaches demonstrates a novel perspective on model enhancement, offering unique insights into improving translation quality without solely relying on pre-existing architectures. The originality in dimensionality reduction lies in the tailored approach for Arabic text. While dimensionality reduction techniques are not new, the paper introduces a specific method optimized for Arabic word embeddings. By employing independent component analysis (ICA) and a post-processing method, the paper effectively reduces the dimensionality of word embeddings while preserving semantic information which has not been investigated before especially for MT task.

Drawing on the politeness theory, this study explored the dubbing of English children's cartoons into Modern Standard Arabic, focusing on negative face speech acts such as compliments, suggestions, rejections, and suggestions in rejections.

Through quantitative and qualitative comparative analyses, this study investigated the influence of cultural differences on the selection of dubbing strategies, analyzed the politeness strategies employed in dubbing, and examined the maintenance of pragmatic equivalence in translated speech acts. Data were collected from the cartoon series “The Loud House” from popular digital platforms Shahid and Netflix, chosen for their regional (the Arab world) and global popularity, respectively.

The findings revealed shifts and adaptations in the dubbing process influenced by linguistic and cultural factors, particularly emphasizing social norms and power dynamics (e.g. respect for elders) prevalent in the Arab culture. Translators strategically employed politeness strategies, such as off-record and bald on-record strategies, to ensure cultural appropriateness and maintain the integrity of familial relationships depicted in the content. The comparative analysis of translators' approaches on Shahid and Netflix platforms highlights significant differences in the translation of negative face-threatening act scenarios. Netflix shows a tendency to adjust politeness pragmatic levels to better suit the preferences of the target audience, indicating a proactive effort to enhance viewer engagement through effective translation strategies. In contrast, Shahid adopts a more conservative approach, maintaining a consistent politeness level in its translations, which aligns closely with cultural norms and expectations. This comparative analysis underscores the divergent translation strategies employed by the two platforms, with Netflix prioritizing adaptation to audience preferences and Shahid favoring preservation of established politeness levels. The cultural social factors that influenced the translation and adaptation of face threatening acts can be grasped by considering elements such as social distance, power dynamics, and relative imposition. In Arabic culture, these factors play a crucial role in determining the social dynamics and interactions between individuals. The observed patterns of shifts and adaptations in the translation of face threatening acts on Shahid and Netflix reflect the nature of stratified Arabic societies where communal relationships are defined by factors like family ties, elder-younger dynamics, and friendships. These cultural factors shape the social norms and expectations surrounding face threatening acts, which, in turn, influence the translation choices made on these platforms.

The study has several limitations that need acknowledgment. Firstly, the sample size was limited, focusing on the dubbing of only two versions of an animated series and four seasons, comparing translations across two platforms. Additionally, there were challenges related to the accessibility and availability of data, with limited information about the translators involved in the dubbing process. Furthermore, the study's focus on specific types of translations may have overlooked other important factors influencing translation practices. Lastly, regional variations in translation practices within the Arab world were not considered. Despite these limitations, the study provides valuable insights into translation practices in animated series on Shahid and Netflix platforms, highlighting the need for future research to address these limitations and provide a more comprehensive understanding of translation practices in the Arab world.

The findings of this study have significant practical implications for a wide range of stakeholders, including translators, educators, textbook writers, and even non-specialists. For translators, especially those working in the field of audiovisual translation, understanding the nuances of politeness strategies and their application in different cultural contexts is essential. The findings suggest that translators should be sensitive to cultural and social factors when translating negative face-threatening act scenarios in cartoons. By making strategic decisions to adapt translations to align with audience preferences while still preserving cultural appropriateness, translators can enhance viewer engagement and ensure that the translated content resonates effectively with the target audience. Educators and textbook writers can also benefit from these findings by incorporating insights into politeness theory and cross-cultural communication into their curriculum and materials. By teaching students about the importance of politeness strategies in translation and providing them with practical examples and exercises, educators can help prepare the next generation of translators to navigate complex linguistic and cultural challenges in their work. Additionally, non-specialists, such as viewers of translated content, can gain a better understanding of the intricacies involved in translation and the impact of cultural factors on communication. By being aware of these issues, non-specialists can become more discerning consumers of translated media and appreciate the skill and artistry involved in the translation process.

This study underscores the cultural competence and sensitivity of Arab translators in navigating linguistic and social dynamics, contributing to the creation of culturally resonant and entertaining content for young Arabic-speaking audiences.

To explore the interplay between human translators and AI tools, focusing on tertiary students' perceptions in the context of Portuguese-Chinese translations in Macao.

This research employed a mixed-methods approach. Quantitative surveys were complemented by qualitative responses. Qualitative class observations (participant and non-participant) and autoethnography further enriched the insights. Participants included undergraduate and postgraduate students in translation studies from the Macao Polytechnic University.

The data revealed a dual perspective: appreciation for AI’s efficiency contrasted with concerns about its potential to overshadow human touch in translations, especially in cultural nuances. Views on integrating AI into curricula were diverse, but a balanced, synergistic approach between human expertise and AI efficiency emerged as a common theme.

This study offers a fresh perspective by integrating various methodologies, capturing both statistical and experiential insights on the evolving relationship between AI and human translation efforts in academia.