Models of sound events produced by simple physical phenomena

Starting date
October 1, 2001
Duration (months)
Computer Science
Managers or local contacts
Rocchesso Davide

Recent developments in information technologies have fostered new forms of communication and human-machine interaction. In particular, new communication channels are being experimented with increasing interest, and they are integrated together with more traditional channels in such a way that a genuine multimodal communication can be established. Within this context, sounds will play an important role in information exchange between machines and individuals, especially in the "ubiquitous computing" scenario, where a multiplicity of objects can interact to process information.
As compared to visual messages, sounds have some advantages: They do not required focused attention, they occupy the space without suffering from obstacle occlusion, and they allow to convey multiple information streams that can be segregated by a human listener or, conversely, the listener is capable to evaluate many streams as a whole.
This research is intended to provide models of sound events generated by interactions of objects, and these models should only deal with the perceptually salient aspects of these phenomena, so that simple interactions can translate into simple and efficient models.
The project belongs to the international context of researches in multimedia systems and human-computer interaction.
Research on sound events and communication by non-verbal messages is still in its infancy, even though important international initiatives demonstrate an increasing interest in this area. For instance, we mention the International Community for Auditory Display (ICAD), an international forum recognised by the National Science Foundation (NSF) where subjects such as "sonification" or "auditory display" are regularly discussed in meetings held in Europe and USA. Nowadays, there are several applications that take advantage of auditory interfaces, in such diverse fields as computer-assisted surgery, control of comples systems, analysis of massive datasets. An important role in defining the state of the art of sound models is being played by the European COST-G6 Action on Digital Audio Effects (DAFx), at its third year of activities, where the partner involved have found a fertile ground to exchange ideas, experiences, models, algorithms, and applications. The activities of DAFx will have its apex in the Conference on Digital Audio Effects (DAFx-00) to be held at the University of Verona on december 7-9, 2000. Through this conference the italian research intends to affirm its leading role in the field in the years to come.


C.N.R. Consiglio Nazionale delle Ricerche
Funds: assigned and managed by the department


Research facilities