|Multilingual Natural Speech Technology|
This research line aims at advancing and developing spoken language processing technology in two major research areas: audio digital libraries and speech translation.
Research is focused on fundamental issues as speech recognition for unconstrained language, ``atypical'' speakers, e.g. children, audio content indexing, cross-language spoken document retrieval, and statistical speech translation.
Research is mostly carried out within national and international projects and in co-operation with
world-class laboratories. Progress is systematically measured through widely accepted benchmarks and by taking part to international evaluations.
Open positions for PhD Students
in collaboration with the
International Graduate School in ICT at the University of Trento.
Phd courses for the academinc year 2002-2003.
Open positions for undergraduate
Tesi di tesi laurea o stage.
Organization of a Cross-Language Spoken Document Retrieval track at CLEF 2003.
Coordination of the CSTAR III 2003 Evaluation Campaign
How to reach us.
Schedule of seminars