You are here: Home / Projects


Overview of past and present R&D projects
Ongoing European projects:
SIIP - Speaker Identification Integrated Project [2014-2018], This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement no 607784,
Malorca - Machine Learning of Speech Recognition Models for Controller Assistance [2016-2018], Horizon 2020,  Call H2020-SESAR-2015-1, Topic Sesar-02-2015 
MuMMER -  MultiModal Mall Entertainment Robot [2016-2020], Horizon 2020,  Call ICT-24.a-2015
SUMMA - Scalable Understanding of Multilingual MediA [2016-2019], Horizon 2020, CallICT-16.a

Ongoing industrial projects:
Ongoing (Swiss) national projects:
Past European projects:
M4 (HLT) – Multi Modal Meeting Manager
AMI(DA) (FP6) – Augmented Multiparty Interaction (with Distant Access)
DIRAC (FP6) – Detection and Identification of Rare Audio-visual Cues
TA2 (FP7), "Together Anywhere, Together Anytime",
Eureka and the European Community under the project “D-Box: A generic dialog box for multilingual conversational applications”, 

Past national projects:
BARRANDE – ''Codade de la parole a tres bas debit independent de la langue'', France.
GRANT AGENCY – Voice technologies for support of information society, Czech Republic.
TAO-CSR, "Task Adaptation and Optimisation for Conversational Speech Recognition",  Innovation Promotion Agency CTI
IM2 (NCCR), Interactive Multimodal Information Management,
Past industrial projects:
Qualcomm, Inc. – Speech and audio coding, USA.
CESNET – Recognition of Keywords and Actions in Meeting

Qualcomm, Inc. – Aurora: Advanced DSR Front-end, USA

Samsung, Inc. (South Korea), "Domain Adaptation Using SubSpace Models"

Swiss Federal Department of Defense, Civil Protection and Sport (armasuisse) - the RECOD project

Samsung, Inc. (South Korea), "Multi-Lingual and Cross-Lingual Adaptation for Automatic Speech Recognition"