Thursday, 12 June 2014 09:18

Panel 11: Pedagogical Approaches to Computer-Assisted Translation Tools

Pedagogical Approaches to Computer-Assisted Translation Tools
Marileide Dias Esqueda, Federal University of Uberlândia, Brazil
Érika Nogueira de Andrade Stupiello,São Paulo State University, Brazil

Considering the diversified scenario in which technology is being taught in undergraduate level in institutions around the world, with some of them having already fully integrated the teaching of tools to their curriculum, whereas others, facing limitations in personnel and infrastructure, are only beginning to include technology in their courses, this panel proposes to address theoretical and practical perspectives of the use and teaching of translation tools at undergraduate level. The objective of this panel is to bring together those concerned about the impact of technology in training of translators, the gap between academia and the market and the impact of new trends, such as crowdsourcing, in translation training.

For informal enquiries: [marileide_esquedaATileelDOTufuDOTbr]

Photo Marileide Esqueda

Marileide Dias Esqueda is a professor at the Federal University of Uberlândia (UFU), where she teaches translation theory and practice in English and Portuguese. She holds a Master's Degree and a Ph.D. in Translation Studies from the State University of Campinas (Unicamp). She is member of the research group Translatio (UFU), sponsored by the National Council for Scientific and Technological Development (CNPq Brazil). Her research areas are pedagogy of translation, and translation tools. She has published articles in edited volumes and journals.

 

Photo.Érika StupielloÉrika Nogueira de Andrade Stupiello is a professor at the São Paulo State University (Unesp), where she teaches translation practice in English and Portuguese. She holds a Master's Degree and a Ph.D. in Translation Studies from the São Paulo State University (Unesp). She has worked as a sworn translator and interpreter for English and Portuguese since 2000. She is member of the research group Multidisciplinary Approaches on Translation (Multitrad), sponsored by the National Council for Scientific and Technological Development (CNPq Brazil), and has been developing research on translation tools, translation ethics and localization. She has published and presented her work both in Brazil and abroad.

 

See other thematic panels

SESSION PLAN

This panel is not divided into thematic sessions.

PANEL STRUCTURE

INTRODUCTION 20 minutes for – Teaching Translation Technology Tools at Undergraduate Level: Challenges and Perspectives

Érika Nogueira de Andade Stupiello and Marileide Dias Esqueda

PAPER 1 – 20 minutes for presentation + 10 minutes for discussion

Title: Integrating Translation Technologies across the Curriculum: A call for a more critical approach to translation tools

Speaker: Matthieu LeBlanc, Department of Translation and Modern Languages at the Université de Moncton (Moncton, New Brunswick, Canada)

Abstract: While translation technologies are now an integral part of university-based translator-training programs, much of the training on technologies is concentrated in specific courses – or in one specific course – devoted to computer-assisted translation (CAT) tools rather than integrated throughout the curriculum, i.e. in general/specialized translation courses or other more theoretical courses. Some have stressed the importance of integrating tools across the curriculum given their ubiquity in the professional workplace and the considerable changes they have brought about to the translation process. Others have suggested that we need to train translators to become critical users of technologies.

In this paper, I will explore the relationship between translation technology training offered in Canadian universities and the integration of junior translators into the workplace. I will draw on qualitative data collected during a three-month ethnographic survey conducted in three different translation firms and services located in Canada. As part of a larger study focusing on translation technologies and professional satisfaction, junior translators were surveyed, via semi-directed interviews, in order to assess their use of technology in the workplace and the relevance of the training they received in university. Senior translators, who are called upon to supervise the work of junior translators, were also interviewed. Translators were also observed at work, at their workstations in order to obtain a better picture of the nature of the work and the use of CAT tools, among other things.

Data analysis revealed that junior translators were generally satisfied with the technology training they received in university and adapted well to tools they were required to use at work. Senior translators confirmed that juniors were indeed sufficiently tech-savvy and certainly not against the use of tools. That being said, nearly all junior translators were surprised to discover the extent to which their work revolved around certain tools, more specifically translation memory (TM) software. In fact, while they did understand how TMs worked and had used such tools in their translation technologies course or even during their internships, they were not aware of the significant role they play in the overall translation process, in establishing productivity requirements and in limiting their control over the target text. In the end, the majority of junior translators felt that they lacked a certain critical awareness of technologies and of their impact on administrative and business practices.

In this paper, I will look at ways that translator-training programs could better integrate CAT tools throughout the curriculum and, more importantly, help students think more critically about the central role played by translation technologies in the professional world.

Bionote: Dr. Matthieu LeBlanc is an associate professor of translation within the Department of Translation and Modern Languages at the Université de Moncton (Moncton, New Brunswick, Canada). He holds a B.A. in Geography & History, an honors B.Tr. in English-French translation, an M.A. in Translation and a Ph.D. in Sociolinguistics. A former professional translator, Dr. LeBlanc now teaches translation, and conducts research in translation and sociolinguistics. His current research focuses on the translator's status and translation practices in an increasingly automated working environment. He has published in the fields of translation, sociolinguistics and language planning, and presented conferences in several countries.

PAPER 2 – 20 minutes for presentation + 10 minutes for discussion

Titile: Applied Research Projects: bridging the gap between research and practice. A case study in MT evaluation.

Speaker: Sandrine Peraldi, ISIT (Paris-France)

Abstract: This paper aims at describing the implementation and the results of an Applied Research Project in the field of machine translation and post-editing. Applied Research Projects are genuine professional projects commissioned by specific companies or research centers (in specialised translation, communication or terminology management) and carried out by Master students under the supervision of researchers and/or experts of the translation industry. These projects enable our students to gain invaluable professional experience by working closely with those companies and meeting their specific industry needs, while developing many research and technological skills.

More specifically, the analysis presented in this paper was initiated by a translation company specialising in regulated financial information. It consisted in evaluating the efficiency of a combined approach of machine translation (MT) and computer-assisted translation (CAT) tools in the financial field. Translation companies specialising in finance are indeed often required to translate extremely high volumes of texts at very short notice, while providing high quality translations to be approved by regulatory authorities. They are thus compelled to streamline as much as possible the translating process, while reducing the number of translators working on the same document as it can affect its quality and its terminological and stylistic coherence.

Despite recent and significant advances in MT (due in particular to the use of hybrid engines that combine linguistic and statistic approaches), computer systems, depending on the domain and the type of texts to be processed, still give rise to relatively poor quality translations, that require an uneconomically large amount of post-editing efforts. On the one hand, the syntactic complexity and the terminological density of the financial field can result in a large number of non-sense phrases, misinterpretations as well as stylistic missteps within the target translation that usually cannot be dealt with through a modification of the syntactic rules or the integrated dictionaries. On the other hand, the high degree of redundancy that characterizes those documents makes CAT tools particularly suitable for integrating machine translation to translation memories. Therefore, our objective was to determine whether the combination of machine translation and computer-assisted tools could offer a credible alternative to human translation from a qualitative and economical perspective.

A three-step methodology was thus implemented. Two different corpora consisting of financial and legal texts were first translated using a specific CAT/MT tool. Secondly, a typology of errors was built by classifying and analyzing the translation segments generated. Finally, this hybrid method was evaluated in terms of time, money savings and quality as compared to human translation. Special attention was paid to the influence of the MT on the post-editor translation choices, during the proofreading stage.

We therefore propose to describe in this paper the different processing steps of our analysis and the final results of this research project, while highlighting the many advantages of such pedagogical projects that enable to gear research and professionalization among Master students.

Bionote: Research Director at ISIT, Sandrine Peraldi holds a PhD in Terminology and Corpus Linguistics (University Paris Diderot). Dr. Peraldi has been in charge of Research projects and European Projects at ISIT since 2006. She lectures in Terminology, Corpus Linguistics, CAT tools and Machine translation. She also supervises Masters research dissertations. As a member of the CRATIL and the CLILLAC-ARP (Paris 7) research centres, she has published more than 30 research articles in the field of onto-terminology, discourse analysis, semantics and machine translation. She is Editor in chief of the Bulletin du CRATIL, the scientific journal of ISIT's research centre.

PAPER 4 – 20 minutes for presentation + 10 minutes for discussion

Title: Exploring the Pedagogical Potential and Challenges of Interactive Translation Dictation

Speaker: Julián Zapata, University of Ottawa (Ottawa, Canada)

Abstract: Our presentation will explore interactive translation dictation (ITD) from a pedagogical viewpoint. ITD is defined as a translation technique that involves mainly voice interaction with multimodal interfaces (MIs) equipped with voice recognition (VR) technology, throughout the entire translation process, namely during preparation, production and revision. Examples of commercially available MIs include smartphones, tablets and touchscreen computers, which are primarily voice- and touch-enabled.

VR is a technological application that gives a machine the ability to recognize and process human voice and speech. Research on VR technology dates back to the early stages of computing in the mid-20th century. Today, after decades of research and development in the field, VR systems are available in several major languages and widely used in a variety of applications, not only for obtaining automatic transcriptions of speech, but also for issuing voice commands to the operating systems in desktop computers and MIs. This technology has the potential to become one of the most efficient, cost-effective and ergonomic applications in the near future for translation professionals, but significant technical and pedagogical challenges still need to be addressed.

In this presentation, we will first provide an overview of the evolution of VR technology and the extent to which it has been explored and used in translation practice and teaching. Secondly, some of the current challenges and limitations of this technology will be described while lending support to the idea of integrating sight translation, translation dictation and VR courses to translator training programs as a partial solution to the challenges. Indeed, as the technology improves, universities play an increasingly crucial role in efficiently integrating VR technology to the translator's toolbox. However, in this integration, translators will have to learn to dictate efficiently; they will have to adopt entirely new translation techniques. To wrap up our presentation, we will present and discuss, from a pedagogical perspective, the preliminary results of an empirical study on ITD carried out within the framework of our doctorate in translation studies (in progress), and will outline avenues for future research.

Bionote: Julián Zapata holds an Honours B.A. in English-French-Spanish translation and a M.A. in translation studies from the University of Ottawa, where he is currently pursuing his doctoral degree. His research interests include multimodal interaction, speech technologies, translation dictation and translation technologies. He is also a professor of translation (English-Spanish) and a teacher and research assistant in translation technology, and has collaborated in a number of projects related particularly to computer-aided translation tools and translator training.

PAPER 4 – 20 minutes for presentation + 10 minutes for discussion

Title: 'Locating' Mobile Localisation into the Translator Training Curriculum

Speaker 1: María del Mar Sánchez Ramos, University of Alcalá (Alcalá de Henares, Madrid, Spain)

Sepaker 2: Lucía Morado Vázquez, Department of Translation Technology at the Faculty of Translation and Interpretation, University of Geneva (Geneva, Switzerland)

Abstract: In the last years, the localisation industry has become a growing market as a means of digital communication, where localisation is defined as the linguistic and cultural adaptation of digital content to the requirements of a specific market or 'locale'. This professional area offers a wide range of possibilities for translators as translation is an essential part of the localisation process. Nevertheless, our undergraduates need technical expertise in order to fulfil the localisation requirements. Teaching localisation implies applying learning methodologies based upon a conceptual and procedural knowledge so that students can acquire a sufficient number of technical capabilities and competences and to prepare them for the current localisation market place. In terms of products to be localised, such as webpages, software programs or videogames, mobile localisation is emerging as a new area and a new market for translators interested in developing more technical skills and become part of the localisation industry. Mobile localisation has its own technical and cultural implications and it can be considered an isolated area of study. Based upon a theoretical and practical approach carried out at the University of Alcalá (Spain) and University of Geneva (Switzerland), we will describe how we have implemented an effective methodology to teach mobile localisation. We will also discuss the main constraints that this specific localisation area entails: the choice of mobile operating system to be used in our labs and tutorials (ios, android, windows phone...) as each of them can imply different teaching procedures and technical implications; the minimalist nature of mobile applications and the specific language that they use; and the difficulty of obtaining real devices to test our localised applications and the current alternative solutions, such as the use of emulators, which are not always able to replicate all the possible case scenarios (e.g. the use of the accelerometer). Finally, we will explain how Computer-Assisted Translation tools can be adapted to train our students on this particular field and how we have made use of their advanced features in our lessons: such as the creation of ad hoc filters for the recognition of translatable text within the source code of a mobile application.

Bionote Speaker 1: María del Mar Sánchez Ramos is a Lecturer in Translation Studies at the University of Alcalá (Madrid, Spain). She studied at the University of Granada (Spain) and received her Ph.D. in Translation Studies from Universitat Jaume I (Castellon, Spain). She completed her post-doctoral studies on Corpus-based Translation Studies at Centre for Translation and Textual Studies, Dublin City University, Ireland. Her main areas of interest are translator training, corpus-based translation studies, localisation and translation technology. Her research has been developed within different national and international projects (ECPC, GEA, AGORA, FITISPOS).

Bionote Speaker 2: Lucía Morado Vázquez is a researcher and lecturer at the Department of Translation Technology at the Faculty of Translation and Interpretation, University of Geneva, Switzerland, on the areas of localisation, computer-assisted translation tools and information technology. She holds a PhD in localisation at the Localisation Research Centre, at the University of Limerick, Ireland. She holds a BA in translation and interpreting at University of Salamanca, Spain. She has been a voting member of the XLIFF Technical Committee and the XLIFF Promotion and Liaison Subcommittee since its establishment. Her research interests are standards of localisation, localisation training and translation memories' metadata.

PAPER 5 – 20 minutes for presentation + 10 minutes for discussion

Title: Translation Technologies: Promises and Challenges for a Less Commonly Taught Language

Speaker: Mehmet Şahin, Department of Translation and Interpretation, Bilkent University (Ankara, Turkey)

Abstract: Computer-assisted translation (CAT) tools and machine translation (MT) systems are becoming ubiquitous for commonly spoken and taught languages such as English, German, French, and Spanish. Technological competence is already included in the EMT (European Master's in Translation) model of translation competence and it is a required skill for translators and interpreters seeking positions in European institutions. Turkish language, although spoken by a large number of people in the Euro-Asian region, has been a less-commonly taught language in the world and received less attention in terms of CAT tools and MT studies. Despite the imbalance in regard to variety and quality of CAT tools and MT and to volume of studies between commonly taught and spoken languages and Turkish language, the volume of translations is growing each day and turn-around time for translation tasks is becoming less and less for both in a parallel manner. Recent surveys show that there is not much consistency in terms of technology use in translation departments in universities and in translation companies in Turkey. Similarly, not all translation departments have a faculty member to teach translation technologies because most of them come from various related fields such as literary studies, linguistics, and education. Even most graduates of translation studies are not fully competent in technology partly because it is an emerging field within translation studies and partly because of lack of interest. On the other hand, translation students and novice translators show a high level of interest in technology courses and eagerness in integrating technology in their work. All of these factors pose a big challenge for translator trainers, translators, and translation companies. How should we teach technology? What should be required from candidate translators in terms of skills and competence? Can we reach a consistency in terms of required competences? Which components of technology should be included in the translation curriculum? How can awareness and research in translation technologies be promoted among scholars? What is the future of CAT tools and MT in Turkey? This study will try to find answers to these questions through documentary research, surveys to translations students and instructors/scholars and to translation companies.

Bionotes: Mehmet Şahin completed his undergraduate studies in the Department of Translation and Interpretation at Bilkent University where he also received his master's degree in Teacher Education. He pursued his doctoral studies in Curriculum and Instruction with a minor in MA in Applied Linguistics/TESOL at Iowa State University. His research studies during his doctoral studies were mainly on computer-assisted language learning and language technologies. Mehmet Şahin is working as an assistant professor of translation and interpretation at Izmir University of Economics since 2008 and his research interests include translation studies, translation and interpreting technologies, machine translation and translator and interpreter training.

WRAP-UP SECTION – 20 minutes

Overviewing the proposals and fostering new ideas to contribute towards the integration of technologies in the translator's training

Érika Nogueira de Andrade Stupiello – Unesp Campus of São José do Rio Preto SP

Marileide Dias Esqueda – Universidade Federal de Uberlândia MG

Back to top

 

© Copyright 2014 - All Rights Reserved

Icons by http://www.fatcow.com/free-icons