Translation in Transition

Between Cognition, Computing and Technology

The main focus of research at the CBS centre for research and innovation in translation and translation technology (CRITT) has been on developing a methodology for translation process research and developing and studying software for translation and post-editing as well as for the study of cognitive processes underlying these tasks. The use of keylogging and eyetracking, sometimes in combination with cued retrospection, less often with concurrent think-aloud, has given much new insight into how translators, at different levels of expertise, produce translations. Our database with recordings of translation and post-editing events now holds close to 1.000 recordings, which we are happy to share with research colleagues around the world. The main tools in this methodological development and research have been Translog and Translog II, both of which have undergone many changes over the years. In the EU FP6 Eye-to-IT project (2006-2009) a gaze-based application (based on Translog and an integrated gaze-to-word mapping tool) was developed to support reading and translation by displaying relevant interlinear prompts whenever a fixation on a word exceeded a time threshold defined by the researcher or user. Currently, in the context of the EU FP7 CASMACAT project (2011-2014), we are developing an ‘intelligent’ interactive application to support post-editing of machine-translated text, for which much of the ‘intelligence’ is derived from our study of expert human translator behaviour.

Hitherto, most of our experiments have involved keyboard and/or gaze input. Only one major project, called ‘Speaking your translation’, has investigated the use of spoken input. However, with recent advances in automatic voice recognition, we foresee that spoken input may soon become standard, either alone or as part of a multimodal key/eye /speech input solution. In order to tackle this new prospect, CRITT has joined forces with the CBS DanCAST centre, which specialises in speech recognition, in a new research platform called The Bridge.

Translation in transition: between cognition, computing and technology celebrates the end of CRITT’s 5-year ‘world-class’ grant from CBS and the transition to the new, larger research group. The conference will provide a forum for discussion both of how we can get to know more about how human translators exercise their skill cognitively and of how the computer can be made to help human translators, by automatically translating written text, by recognising and translating spoken utterances or –more indirectly– by logging translation events and analysing recorded process data.

We therefore invite papers on such subjects as:

  • translation process research

  • reading processes (especially in translation)

  • writing and post-editing processes

  • intelligent machine translation (MT which learns from/adapts to human users)

  • cognition and translation

  • computational modelling of human translation

  • speech recognition and translation

  • translation expertise, in people and in systems

Important information & dates

  • Keynote speakers:

    • Srinivas Bangalore (AT&T Labs-Research, USA)

    • Annette de Groot (Universiteit van Amsterdam, The Netherlands)

    • Paul Kussmaul (Johannes Gutenberg-Universität Mainz, Germany)

    • Shravan Vasishth (Universität Potsdam, Germany)

    • Luuk Van Waes (Universiteit Antwerpen, Belgium)

  • Scientific committee: Laura W. Balling, Michael Carl, Barbara Dragsted, Peter Juel Henrichsen, Arnt Lykke Jakobsen, Andreas S. Kirkedal

  • Conference fee: 750 Kr. (approx. 100 euros - it includes book of abstracts, coffees and lunches)

  • Conference dinner: January 30, 2013. This conference dinner will be optional and sponsored by CBS at an additional cost of 30 euros

  • Contact: Bartolomé Mesa-Lao (bm.ibc@cbs.dk)

CONFERENCE PANELS

PANEL I – TRANSLATION AND COGNITIVE PROCESSES

  • Hill et al.: Human detection of translation errors in text: unwrapping the dynamic process through eye-tracking

  • Huang et al.: Cognitive process of revision: the behaviours and the motivations behind

  • Hvelplund: The processes of translation

  • Schaeffer et al.: Reading for translation

PANEL II – TRANSLATION AND TRANSLATION THEORY

  • Čulo: Translation machine, machine translation, translation theory – some observations and suggestions

  • Hietaranta: Viewing translation problems as threats: a cognitive approach to misunderstandings in translation

  • Lei: Large-corpus-based cognitive study of translation and interpreting processes: methodological procedures and theoretical implications

PANEL III – TOOLS FOR TRANSLATION PROCESS RESEARCH

  • Carl: Product- and process units in the CRITT translation process research database

  • Serbina et al.: Towards a user-centered platform for analysis of multimodal data

  • Couto Vale: Writing strata and Translog-II: towards an integration with Litterae, LETRA’s annotation tool for translation process data

PANEL IV – APPROACHES TO TRANSLATION PROCESS RESEARCH

  • Ferreira: L1 and L2 translation: a metacognitive analysis of the bilingual subcompetence

  • Schmaltz et al.: Participant tracking in text unfolding: insights from Portuguese-Chinese translation and post-editing task logs

  • Vandepitte et al.: Reading metonymic constructions: an eye-tracking investigation

  • Whyatt et al.: Exploring decision-making processes in translation and paraphrase: a report on the ParaTrans project

PANEL V – TRANSLATION AND POST-EDITING

  • Alves: Tracking post-editing effort from a relevance theoretic perspective: the conceptual/procedural distinction under scrutiny

  • Nitze et al.: Post-editing and translation strategies

  • Guerberof: Productivity and quality in the post-editing of outputs from translation memories and machine translation

PANEL VI – SPEECH AND TRANSLATION

  • Barbin et al.: Translation technology and learner performance: professionally-oriented translation quality assessment with three translation technologies

  • Kirkedal: Automatic phonetic transcription for Danish speech recognition

  • Liyanapathirana: Integrating automatic speech recognition and machine translation for better translation outputs

  • Mesa-Lao: Post-editing through speech recognition: a feasibility study with post-editor trainees

INVITED SPEAKERS

  • Bangalore: Machine assisted human translation as a multimodal dialog (PDF)

  • de Groot: Language control, executive control, and processing control in bilinguals and translators/interpreters (PDF)

  • Kussmaul: Observing and interpreting the human translation process (PDF)

  • van Waes: Fluency revisited: A multi-dimensional approach to text composition (PDF)

  • Vasishth: Reading and sentence comprehension processes (PDF)