Tharindu Ranasinghe, University of Wolverhampton

TransQuest: Translation Quality Estimation with Cross-lingual Transformers

1 October 2020

Abstract:

High-accuracy translation quality estimation (QE) that can be easily deployed for a number of language pairs is the missing piece in many commercial translation workflows. Even though there are many systems that can do QE, majority of these methods work only on the language pair they are trained on and need retraining for new language pairs which can be usually computationally expensive and difficult especially for low-resource language pairs. As a solution, in this presentation, we introduce TransQuest – a simple QE framework based on cross-lingual transformers. TransQuest outperforms the current state of the art quality estimation methods like DeepQuest and OpenKiwi. This is also the winning solution in recently concluded WMT 2020 sentence-level Direct Assessment shared task, winning all the language pairs with the multilingual track too. 

Dr Stephen Doherty, The University of New South Wales, Sydney, Australia

Eye movements, cognitive load, and human-computer interaction with translation and interpreting technologies

2 December 2020

Abstract

Technological advances have led to unprecedented changes to translation and interpreting (see Doherty, 2016), chiefly in how we access and use translation and interpreting technologies for a diverse and growing range of professional and personal activities. Previous empirical research on translation and interpreting technologies has yielded a wealth of evidence to advance our understanding and usage of these technologies in addition to making them more visible and accessible. Of particular value amongst this growing body of work is the use of eye tracking in exploring and understanding the psychological and cognitive aspects of translation and interpreting technologies by analysing our eye movements as we interact with these technologies and use their outputs.

In this paper, I will consolidate this work by presenting a critical review of empirical studies of translation and interpreting technologies which have employed eye tracking, including my own recent work in the Language Processing Lab at the University of New South Wales. I will categorise previous research into areas of application, namely: computer-assisted translation tools, quality assessment of machine translation, post-editing machine-translated output, audio-visual translation, and remote interpreting. In closing, I will discuss the strengths and limitations of eye tracking in such studies and outline current and future research.

Suggested background reading:

Doherty, S. (2016). The impact of translation technology on the process and product of translation. International Journal of Communication, 10, 947–969.

Speaker’s bio

I am Associate Professor in Linguistics, Interpreting, and Translation, and lead of the HAL Language Processing Research Lab at UNSW. With a focus on the psychology of language and technology, my research investigates human and machine language processing using natural language processing techniques and combinations of online and offline methods, mainly eye tracking and psychometrics. My research has been supported by the Australian Research Council, Science Foundation Ireland, the European Union, the Federal Bureau of Investigation, the National Accreditation Authority for Translators and Interpreters, New South Wales Ministry of Health, Enterprise Ireland, and a range of industry collaborations. As a Chief Investigator, I have a career total of $1.5 million competitive research grants. 

Prior to my appointment at UNSW Sydney (2014), my doctoral (2008–2012) and post-doctoral research positions (2012–2013) were funded by Science Foundation Ireland and supervised by Prof Sharon O’Brien, Prof Dorothy Kenny, and Prof Andy Way at the CNGL Centre for Global Intelligent Content in Dublin City University, a multi-million euro, cross-institutional centre now known as the ADAPT Centre for Digital Content Technology. My subsequent post-doctoral position (2013–2014), supervised by Prof Josef Van Genabith, was based in the School of Computing and the National Centre for Language Technology at Dublin City University as part of the QTLaunchPad project, a €2.2 million project funded by the European Union through its Seventh Framework Programme (FP7) for research and technological development.

Dr Juan José Arevalillo

Quality standardisation in language industry

27 November 2020

Abstract:

Quality in the language services industry is a very blurred term, but omnipresent in all activities from the moment a customer requests a translation service to the delivery and final closing of the project. In this process everything is measured and compliance with all requirements is usually a guarantee of success with the customer. In addition, there are numerous quality standards under ISO’s umbrella covering different services and aspects in this industry, which are applied on a daily basis and also form the basis of numerous academic programmes. Knowing this environment is essential for the future professionals so that they can know where they fit into the process and how to behave and act in that process.

Bio:

PhD in Translation by University of Malaga, MA in Specialised Translation by the Institute of Modern Languages and Translators by Madrid Complutense University and BA in English Language and Literature by Madrid Complutense University.

In translation industry since 1980, he is the Managing Director at Hermes Traducciones y Servicios Lingüísticos. Previously worked as a freelance translator and as a language specialist and localiser in Digital Equipment Corporation.

A lecturer on Translation at Alfonso X University (Madrid) and International University of Valence (Spain), he is also the professional advisor for future graduates in the former university. He works with other Spanish high-education centres such as Autonomous University of Madrid, Autonomous University of Barcelona and ISTRAD of Seville.

Formerly Vice-president and Treasurer of the European Union of Associations of Translation Companies (EUATC), now he is the EUATC Youth Ambassador to try to cover the gap between university and industry and help new graduates join professional world. He is also the Chairman of the Spanish Association of LSPs (ASPROSET). Chairman of Spanish Committee for Translation Services at UNE (Spanish Standardisation Association), and one of the creators of EN-15038 and ISO-17100 standards. He is also a member of ISO TC37 Committee for Translation Services.

Technologies for Translation and Interpreting: Challenges and Latest Developments

Dr Mark Shuttleworth

Free translation memory tools: a comparison of some well-known systems’.

25 Novemeber 2020

Abstract

‘The use of translation memory tools is now fairly well embedded within the profession. While many translators are obliged to use one or other well-known system, others who are able to choose for themselves are perhaps confused by the sheer choice of systems that are available. In this talk I will demonstrate Memsource, Wordfast and Matecat and attempt to answer the following two questions: 1) to what extent does a free tool provide you with the functions that are needed to work at a professional level and 2) what are the strengths and weaknesses of each of these three systems?’

Speaker’s bio:

Mark Shuttleworth has been involved in translation studies research and teaching since 1993, at the University of Leeds, Imperial College London, University College London and, most recently, Hong Kong Baptist University. His publications include the Dictionary of Translation Studies, as well as articles on metaphor in translation, translation technology, translator training, translation and the web, and Wikipedia translation. More recently he has become interested in the use of digital methodologies in translation studies research. His monograph on scientific metaphor in translation, Studying Scientific Metaphor in Translation, was published in 2017 and he is currently working on a second edition of the Dictionary.

Keynote speaker engagements have included translation studies conferences in Poland, China and Malaysia. He has also addressed industry conferences in the UK, Italy and Brazil on the subject of translation technology and has provided training in the same area in Spain, Italy, Portugal, Finland, Tunisia and Malaysia.

Mark Shuttleworth is a fluent speaker of Russian, German, Polish and French and has some knowledge of a number of other languages including some Chinese. As and when time permits he is active as a translator