This demo shows the FBK Acoustic Event Detection and Classification System. The real-time prototype is able to recognize a set of 12 different acoustic events. Some examples of events are: door knocking or slamming, phone ringing, laughs, coughs, and obviously speech.
In this demo, the classification is based on the input from a given microphone, while the location of the source is based on 14 microphones. The classification is based on Hidden Markov Models. Event accuracy is around 95%.
Possible applications: smart rooms, surveillance, lectures, meetings, domotics, robotics, security, acoustic scene analysis.