Departments and Units

Projects

The Department of Translation Technology (TIM) leads and participates in several research projects. You can consult the ongoing projects as well as the completed projects.

On-going projects

Spoken Language Translation of Dialogues in the Medical Domain

BabelDr is a spoken language translation system for the medical domain, developped in collaboration with Hôpitaux Universitaires de Genève (HUG), Geneva's largest hospital. Its main focus is precision and less-resourced languages.


Funded by: Mimosa project, HUG's private foundation and Innogap grant.

CALLector, which officially started on April 1 2018, is funded by the Swiss National Science Foundation under its COST program, and the project is closely linked with the enetCollect COST project. The overall goal of CALLector is to create a platform, structured as a social network, for crowdsourced construction and use of CALL content.


Funded by: Swiss National Science Foundation (SNSF), COST program.

Collaborative Post-Editing Corpus in Pedagogical Context

COPECO is a joint project between Geneva University and Liège University, with three main objectives: 1) to collect post-edits produced by students and teacher corrections, 2) to build an open-source student post-editing corpus and 3) to help systematise the task of translation error annotation. It provides translation teachers with an online post-editing platform, designed to help them to annotate student post-editing tasks using a shared or personalised annotation scheme.


Funded by: projet COINF 2020, UNIGE.

Effective health translation and communication strategies among multicultural populations lie at the heart of efforts to combat physical and mental diseases in multicultural societies like Australia and Switzerland. There is a persistent lack of standardised criteria and benchmarks for the development and evaluation of healthcare and medical translation resources and computerised translation systems. The lack of international standards and guidelines have impeded the effective adoption and implementation of health translation resources and technologies in healthcare research and clinical settings. Our project represents a first attempt to tackle this persistent, costly issue, which has become increasingly urgent in multicultural Australia and the refugee crisis in European countries like Switzerland. 


Funded by: USyd - UNIGE - Partnership Collaboration Awards (PCA).

Subtitling of Swiss German into Standard German: Automatic Post-Editing

This project aims to develop, in collaboration with SRF, an automatic post-editing system to improve the quality of automatic German subtitling of Swiss German television programs. It has three main objectives. The first one is to compare different methods of automatic post-editing for this task, the second one is to collect the opinions of users on this type of transcription in order to generate subtitles that are as accessible as possible, and the third one is to share written and oral resources to promote research on Swiss German in the field of media.


Funded by: IMI (Initiative for media innovation).

PRojection du langage Oral vers des unités PICTOgraphiques

PROPICTO is a collaborative project between the University of Geneva and the University of Grenoble Alpes which intends to develop a line of research on Augmentative and Alternative Communication (AAC). The main objective of PROPICTO is to create systems that automatically transcribe French speech into pictographic form, and thus to facilitate communication access in the medical field (especially with patients who do not speak the same language as the practitioner) or in the area of disability (and so to promote a better interaction with people having a cognitive impairment).


Funded by: French National Research Agency (ANR) and Swiss National Science Foundation bilateral project.

UNI-ACCESS

Accessibilité des sites web des Hautes écoles

This project aims to identify current accessibility barriers in university websites. It proposes specific linguistic and technical solutions to integrate easy-to-understand language and sign language into university websites.


Funded by: Swissuniversities.

Completed projects

European projects

  • enetCollect - European Network for Combining Language Learning with Crowdsourcing Techniques
  • ACCEPT - Automated Community Content Editing PorTal
  • DicoPro - On-Line Dictionary Consultation For Language Professionals On Intranet.
  • DiET - Diagnostic and Evaluation Tools for Natural Language Applications.
  • EAGLES I - Expert Advisory Group on Language Engineering Standards / Evaluation Working Group Final Report I.
  • EAGLES II - Expert Advisory Group on Language Engineering Standards/Evaluation Working Group Final Report II.
  • eCoLoRe - Creating Shareable and Renewable eContent Localisation Resources to Support ICT Training for Translators.
  • GRAAL - Grammars which are Reusable to Automatically Analyze Languages.
  • ISLE - International Standards for Language Engineering.
  • LCG - Learning Computational Grammars.
  • MeLLANGE - Multilingual eLearning in LANGuage Engineering.
  • MLCC - Multilingual Corpora for Coorporation.
  • MULTEXT - Multilingual Text and Tools.
  • Parmenides - Ontology-driven Temporal Text Mining on Organisational Data for Extracting Temporal Valid Knowledge.
  • TEMAA - A Testbed Study of Evaluation Methodologies: Authoring Aids.
  • TransRouter - Translation Router.
  • TRANSTERM - Use of Terminological Data in NLP Systems.
  • TSNLP - Test Suites for Natural Language Processing.

Projects of the Swiss National Science Foundation (SNSF)

  • A Swiss Platform for Controlled spoken dialog applications - The project proposes to extend and stabilise the Regulus platform so as to make it suitable for adoption as a general Swiss platform for spoken language system construction in academia and industry.
  • A generic platform for CALL based on speech translation - We leverage Geneva's Regulus toolkit and other SNSF-sponsored work on speech recognition and translation, to develop a generic CALL platform centered on the 'spoken translation game' idea.
  • CALL-SLT 2 - Designing and Evaluating Spoken Dialogue Based CALL Systems
  • DicoEast - SCOPES project for electronic dictionary distribution.
  • MedSLT - A Multilingual Spoken Language Translation System for Medical Domains.
  • Medtag - Free text medical document retrieval by using terminological resources and statistical linguistics.
  • Medtag2 - Free-text medical document retrieval via extraction of complex terms.
  • MTEval - Quality models and resources for the evaluation of machine translation.
  • MTEval II (FEMTI on-line application) - Quality requirements and metrics for the evaluation of MT: analysis and integration of expertise.
  • TransTech - Language Technology for Translators.
  • (IM)2 - Interactive Multimodal Information Management.
  • MT Sublanguage - Definition and Exploitation for Sublanguage Description for MT in a finite domain, part I and II.

Projects of the Ernest Boninchi Foundation

  • Web-Based Deployment of Speech-Enabled Applications for Medicine and Education.
  • Trainslate - Spoken Language Translation to Sign Language.

"Suissetra" federal projects

  • GTF-Rifal - Contribution to the Rifal courses and textual databases.
  • OSIL - Observatoire Suisse des Industries de la Langue (Swiss Observatory of Language Industries).
  • OTIL - Observatoire francophone du Traitement informatique des langues.
  • FLAG - Searching and verification of terms in multilingual documents.
  • SYLEX - Browser for multilingual federal law documents (view demo).

Projects of the French Speaking Community

  • Automatic acquisition of Generative Lexicon elements to increase performances of information retrieval systems (AUPELF-UREF project, FRANCIL network).
  • Alignment of Bilingual and Multilingual Corpora.
  • CRISTAL - Contextes RIches en ConnaissanceS pour la TrAduction terminoLogique
  • LEGERe - Un LExique GEnératif de Référence pour le français (project description in French).

Other Projects


Back to top