Research Projects

IICT is a four-year project with an overall budget of  12.3 million CHF supported by the Swiss Innovation Agency (Innosuisse) under its flagship scheme. The goal of the IICT flagship is to develop information and communication technologies (ICT) in the context of persons with disabilities. The flagship targets five applications: text simplification, sign language translation, sign language assessment, audio description, and spoken subtitles.

SMILE II aims to research and build advanced technology for sign language learning. More precisely, the project builds on the groundwork of the SNSF Sinergia project SMILE I, which dealt with assessment of the manual activity of Swiss German Sign Language (Deutschschweizerische Gebärdensprache, DSGS) in isolated signs produced by L1 users and L2 learners. SMILE II extends this technology to continuous sign language assessment including both manual and non-manual components of signs so that a DSGS learner’s sentence-level production can be assessed in an automatic manner.

This project has at its core research into semi-automatic language sample analysis. The focus is on the development and evaluation of a software which can be used to support speech therapists in the diagnostics of language disorders in children. In a first phase, the goal is to create a corpus with spontaneous speech of children of ages four to six with typical and atypical language development in Swiss German and Standard German. This data is then used to train automatic speech recognition and automatic language sample analysis models. The process targeted is semi-automatic, with a human (i.e., an expert) in the loop.

EASIER aims to create a framework for barrier-free communication among deaf and hearing citizens across Europe by enabling users of European sign languages to use their preferred language to interact with hearing individuals.

In this project, together with CFS GmbH ("capito") in Graz, Austria, we worked on semi-automatic text simplification for German with the aim of producing simplified German at levels A1, A2, and B1 according to the Common European Framework of Reference for Languages (CEFR). The output of our neural machine translation models is postedited by human translators.

The goal of this project was to pioneer an assessment system for Swiss German Sign Language (Deutschschweizerische Gebärdensprache, DSGS) using automatic sign language recognition technology. To achieve this goal, the project used a multidisciplinary framework that followed two strands of research, one on sign language technology and one on sign assessment with a common link to sign language linguistics.

In the Trainslate (=train+translate) project we developed a system that automatically translates German train announcements of the Swiss Federal Railways (Schweizerische Bundesbahnen, SBB) into Swiss German Sign Language. The idea for such a system was suggested to us by Deaf signers in Switzerland.

This two-part project dealt with Swiss German Sign Language. The first part involved sign language machine translation, the second sign language animation assessment, i.e., the evaluation of virtual signers, or 'avatars'.