15:00-18:00
Museo degli Strumenti per il Calcolo – Palazzina A1
Moderator: Gianni del Vecchio
Speakers: Roberto Pieraccini, Mirko Tavosanis, Franco Cutugno, Roberto Basili, Maria Palmerini, Carlo Aliprandi
Language technologies are rapidly changing the way we interact with computer systems. The event takes stock of these fascinating innovations, comparing the views of Italian and international experts who tackle the subject from different angles.
15:00-15:10 // KEYNOTE SPEECH
Introduction
Speaker: Mirko Tavosanis
15:10-15:30 // KEYNOTE SPEECH
Multimodal dialogue systems
Speaker: Franco Cutugno
In this paper I will present some preliminary results of the CHROME project (http://www.chrome.unina.it – PRIN 2005 MIUR), which aims to gather linguistic and multimedia resources to guide the generation of multimodal services in the context of technology development sector at the service of improving the use of cultural heritage. I will describe the phases of collecting multimedia linguistic materials and acquiring specialized texts on history of art and architecture, which are used to develop systems of human-machine interaction based on dialogue and multimodal interaction (gestures + voice). These systems are integrated by studies and architectural surveys, based on recent technologies, leading to the virtual reconstruction of three-dimensional real environments of sites of cultural interest (our project is studying the three Certose of Campania: San Martino in Naples, San Giacomo in Capri, and San Lorenzo in Padula). The systems we are developing are then inserted into three-dimensional virtual environments in which both humans and avatars “move”. Humans “navigate” in the three-dimensional reconstruction of the Certose and the avatars simulate the behaviour of tourist guides that illustrate the characteristics of the works. Humans ask the avatars questions and point out details of the virtual environment, and the system interacts by acquiring multimodal requests and merging the vocal part with the gestural part to give the best possible response to the user’s requests. I will try to introduce all the research aspects and challenges that the pursuit of these objectives involves.
15:30-15:50 // KEYNOTE SPEECH
Natural language, machine learning and robotics
Speaker: Roberto Basili
15:50-16:10 // KEYNOTE SPEECH
Speech recognition problems
Speaker: Maria Palmerini
16:10-17:00 // PANEL
Panel discussion
Speakers: Franco Cutugno, Mirko Tavosanis, Roberto Basili, Maria Palmerini, Roberto Pieraccini, Carlo Aliprandi
17:00-18:00
Demo of Musuem Guide with Google Home