|
|
Oferta de Projecte Final de Carrera:
Activity detection and description is a key functionality of perceptually aware interfaces working in collaborative human communication environments like meeting-rooms or classrooms. In such types of environments the human activity is reflected in a rich variety of acoustic events, either produced by the human body or by objects handled by humans, so acoustic scene analysis may help to detect and describe human activity as well as to increase the robustness of automatic speech recognition systems. Indeed, speech usually is the most informative acoustic event, but other kind of sounds may also carry useful information. For example: clapping or laughing inside a speech, a strong yawn in the middle of a lecture, a chair moving or door noise when the meeting has just started. The proposed PFC will focus on detection of acoustic events that may take place in meeting-rooms or classrooms. The number of sounds encountered in such environments may be large, but here we concentrate on a given set of 10-15 different acoustic events, including speech and music. The work consists of developing and testing alternative algorithms, and also implementing an online version in the UPC's smart room (building D5), using the signals captured by the set of distant microphones available in that room. The main targets of the project are:
The ideal candidate should:
Contacts
Andrey Temko (PhD student) or Climent Nadeu (professor)
TALP Research Center
Department of Signal Theory and Communication
Polytechnic University of Catalonia
Campus Nord, D5, 118
Jordi Girona 1-3,
08034 Barcelona, SPAIN
Tel: +34-93 401 1627
Fax: +34-93 401 6447
E-mail: temko@gps.tsc.upc.es
http://www.talp.upc.es
http://gps-tsc.upc.es/veu/
http://gps-tsc.upc.es/veu/personal/temko/
|
|
Send us any suggestion
| This page was last updated on September 12th, 2005 |