Research Projects
The following research projects have received external funding
This project is a joint effort to improve products and processes linked to translators' interaction with technology. We will examine the efficacy of different working methods and assess new training approaches that can be deployed in the translation workplace. In a context of heated debates on the impact of automation on human labour, our aim is ultimately to empower translators to adopt machine translation in more effective and rewarding ways that can improve their job satisfaction, raise professional standards and boost the multilingual economy.
Collaborators:
The project seeks to integrate evidence from translation product research and translation process research, and to develop an integrated cognitive model of the translating mind based on empirical evidence, which will be validated in a broad range of language combinations.
MEMENTO organizes yearly Boot-camps on empirical Translation Process Research (TPR)
Project partners are:
Prof. Victoria Lai Cheng LEI, University of Macau (UM)
Prof. Defeng LI, University of Macao (UM), China
Prof. Michael CARL, Copenhagen Business School / Renmin University of China (RUC)
Prof. Silvia HANSEN-SCHIRRA, Johannes Gutenberg-Universität Mainz (FTSK), Germany
Prof. Fabio ALVES, Universidade Federal de Minas Gerais (UFMG), Brazil
Dr. Moritz Schaeffer, Johannes Gutenberg-Universität Mainz (FTSK), German
2011-2014: Cognitive Analysis and Statistical Methods for Advanced Computer Aided Translation (CASMACAT) (2.5M Euro)
This project seeks to advance computer-aided translation, and integrate novel models into a new open source workbench, in order to improve productivity of human translators by addressing their needs for the right type of assistance at the right time. An important objective of the CASMACAT project is to gain insight into the cognitive processes involved in human translation. Relying on key logging and eye-tracking, we study translator behaviour in computer-aided translation, and investigate the usefulness of visualisation options in post-editing and interactive translation, for different types of text, for different language pairs, and for translators with different degrees of expertise. The findings of this first stage provide the theoretical background for the CASMACAT work on interactive translation prediction and interactive editing, and are crucial for the development of the adaptable CASMACAT workbench. Based on the cognitive user model, it will anticipate user behaviour and tailor visualisation to the users’ immediate needs. Partners: University of Edinburgh, Universitat Politècnica de València, Celer Soluciones, and Copenhagen Business School. Members from CBS: Arnt Lykke Jakobsen, Michael Carl, Moritz Schaeffer, Ragnar Bonk, Mercedes Garcia Martinez, Nancy Underwood, and Jakob Elming.
Several 'side-projects' have been conducted that were partially co-financed by CASMACAT:
From July 13 to August 15, 2014, the CRITT plans to conduct a Translation Data Analytics (TDA) project.
The aim of the TDA project is to explore and analyse user activity data which is collected in advanced man-machine communication situations. The TDA project will assess and elaborate methods to produce data-driven user profiles, to investigate differences in communication styles, and to identify patterns of user behavior for more and less successful man-machine communication. Data repositories such as the CRITT TPR-DB will be taken as a basis to analyze concrete and specialized forms of professional man-machine communication such as translators behaviour in advanced computer-assisted collaborative production environments.
An introductory PhD summer course on Translation Process Research (TPR) and a one-day workshop will precede the TDA project, in which students get acquainted with peculiarities of the TPR data and data acquisition methods. The TDA project will use exploratory statistical approaches for discovering new features and dependencies in the TPR data, and to formulate hypotheses about the causal relations. Statistical hypothesis tests will be deployed for confirming or falsifying existing hypotheses.
During summer 2013 The Bridge (CRITT + danCAST) plans to conduct an implementation workshop for computer assisted translation (CAT), in which a translator reads a source text on a computer screen and speaks out the translation in the target language, a process called sight translation. This sight translation process is supported by an Automatic Speech Recognition (ASR) and a Machine Translation (MT) system, which transcribe the spoken speech signal into the target text and which assist the translator with partial translation proposals, predictions and completions on the computer monitor. An eye-tracking device follows the translators gaze path on the screen, detects where he or she faces translation problems and triggers reactive assistance.
2011-2013: Networking Grant (DKK 1 million)
Between 2011 and 2013, the Danish Agency for Sciences, Technology and Innovation supported the CRITT with international networking grants, under the following subtitles:
Human Machine Interaction in Translation 1.3.2011 – 31.12.2011
The aim of this network activity was to specify common interests between the three research groups CDAC-Noida/India, LETTRE at Universidade Federal do Minas Gerais, Belo Horizonte/Brazil and CRITT at Copenhagen Business School in the field of Translation Process Research, to discuss and specify colaboration projects of common interest and to draft project proposals for possible future colaborative research project(s).
Post-editing Expertise 1.3.2012 – 31.12.2012
The aim of the network activity “Post-editing Expertise” was to investigate post-editing behaviour from a technological and a human angle. It aimed at discussing and promoting novel forms of post-editing machine translation output, including new technological possibilities as well as human experience with those technologies. The network was extended with groups from NICT/Japan and HKUST/Hong Kong.
Speech and Gaze in Translation 1.1.2013 – 31.12.2013
The network “Speech and Gaze in Translation” extends the 2012 network activity by exploring new modalities of how translation problems can be tackled.
The network organized a large number of meetings, workshops and conferences in Denmark, India, Japan and the USA (see here). International partners of the network were:
Prof. Michael Carl, Copenhagen Business School (CBS)
Dr. Srinivas Bangalore, AT&T Labs-Research, NJ, USA
Prof. RMK. Sinha, JSS Academy of Technical Education, Noida and former Professor & Head, CSE, IIT Kanpur, India
Prof. Fabio Alves, Belo-Horizonte/Brazil
Dr. Eichiro Sumita, Director of Multilingual Translation Laboratory, Project Leader of MASTAR Project of NICT and Visiting Professor of Kobe University Graduate School, Japan.Activities
Prof. Dekai Wu, Hong Kong University of Science and Technology
2009-2011: Reconstruction of Speech Events in Real Time (DKK 4.7 million)
The project aims to rethink the way information is encoded in speech and decoded in hearing. Our formal models are optimized for compatibility with hearing aids, speech technology, and mobile telephony. We use methods from machine learning, numerical analysis, deductive systems, acoustic analysis, and natural language parsing, based on data from real-world and laboratory tasks. The project is funded by the Danish Research Council for the Humanities. Members: Peter Juel Henrichsen (principal investigator), Maya Borges, Ruben Schactenhaufen, Thomas Ulrich Christiansen (Technical University of Denmark), Andrea Corradini (University of Southern Denmark).
2008-2011: The Copenhagen Dependency Treebank (DKK 3.0 million)
This project seeks to build a set of parallel treebanks for Danish, English, German, Italian, and Spanish with 40,000 words in each language, based on a unified annotation scheme for syntax, morphology and discourse. The treebanks will then be used to explore a wide range of research problems in theoretical and computational linguistics. The project is funded by the Danish Research Council for the Humanities. Members: Matthias Buch-Kromann (principal investigator), Iørn Korzen, Henrik Høeg Müller, Martin Haulrich.
2007-2010: The Discontinuous Grammar System
This project seeks to develop dependency-based models of human parsing and translation, exploring a new serial repair-based processing architecture based on local search. The project is funded by the Danish Research Council for the Humanities. Members: Matthias Buch-Kromann.
2007-2009: Comprehension and Text Production in Translation and Interpreting Hybrids (FKK Project)
Using eyetracking, keylogging, voice recognition, and audio recording, this project explored the interaction of comprehension and production processes in hybrid forms such as sight translation, which is hybrid in the sense that it involves production of oral target text from a written source and therefore demonstrates a cross of features from traditional (written) translation and (oral) interpreting. The theoretical aim was to explore the different forms of interaction in closely related translation genres, and the practical aim was to demonstrate the positive and/or negative effects of convergence of the written and spoken modes. Members: Arnt Lykke Jakobsen (principal investigator), Barbara Dragsted, Inge Gorm Hansen. For results, see under Publications by the members of the research team.
Tools and Technologies for the Purposes of Studying Cognition and Translation (EU-FP6 project) This project sought to build an application that can be used at once for studying human translational behaviour and to support practical translation. The application interprets information about a translator's eye and finger movements, provides insight into comprehension processes and text production processes, and also finds out when a translator needs help and provides that help instantly. Partners: University of Tampere, New Bulgarian University, Tobii Technology AB, Technical University Graz, University of Oslo, Copenhagen Business School. Members from CBS: Arnt Lykke Jakobsen (principal investigator), Barbara Dragsted, Michael Carl, Christian Jensen, Kristian Tangsgaard Hvelplund Jensen, Annette Camilla Sjørup. For results, see under Publications. For detailed info on the entire project, see https://cordis.europa.eu/project/rcn/86408/factsheet/en
The following PhD projects have been conducted within CRITT
(2011-2014) Morten Gylling-Jørgensen:
The Structure of Discourse: A Corpus-based Cross-language Study
Project description: The project will determine and describe the upper level for language, i.e. discourse, which involves text relations beyond the syntactic structure. The traditional linguistic research has mainly been concentrated inside syntax, but new development in computational linguistics, e.g. linguistic annotation of corpora, has facilitated studies in larger text units, such as discourse units. A better understanding of discourse does not only lead to a better ability to communicate and to strengthen the cohesion in a text, but it also contributes to a more precise understanding of the structure of texts in various languages. The primary aim of the project is to prove and explain language typological differences in the discourse structure between the two well-known language families Germanic and Romance. This will help improving translation processes and foreign language pedagogy.
(2009-2013) Annette Camilla Sjørup:
The process of translating metaphors: an analysis of the processes and behaviours of the translator when translating metaphors
Project description: The project uses eye-tracking and keylogging technologies to seek a pattern in gaze and typing activity for the translation of metaphorical expressions. The project hypothesises that gaze times will be longer for metaphorical expressions than for non-metaphorical expressions. The baseline measure used is total gaze time/character to enable comparisons between segments of varying length.
(2008-2011) Kristian Tangsgaard Hvelplund Jensen:
Eye movements, cognition and translation
Project description: Using a combination of eye tracking and keylogging technologies, this project aims at exploring the distribution of attention during translation from English to Danish. The experimental texts used in the study differ with respect to their levels of complexity, and a time constraint was introduced in a number of tasks. More specifically, it aims at investigating (1) the frequency with which attention shifts between source text and target text during translation, and (2) the duration of the attentional segments based on segment type. Completed 2011.