BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Variational Inference with Bayes Net Fragments for Beat Tracking a
 nd Rhythm Recognition - Charles Fox
DTSTART:20070601T143000Z
DTEND:20070601T160000Z
UID:TALK6976@talks.cam.ac.uk
CONTACT:Taylan Cemgil
DESCRIPTION:It is useful for music perception and automated accompaniment 
 systems to perceive a music stream as a series\nof bars containing beats a
 nd rhythm patterns. We present a method combining variational Bayesian inf
 erence in network fragments with a blackboard system for simultaneous beat
  tracking and rhythm pattern recognition in the domain of semi-improvised 
 music. This is music which consists mostly of known bar-long rhythm patter
 ns in an improvised order\, and with occasional unknown patterns. We assum
 e that some lower-level component is available to detect and classify onse
 ts. Model posteriors provide principled model competition\, and the system
  may be seen as providing a Bayesian interpretation of agent-based and bla
 ckboard systems. \n\n
LOCATION:Engineering Department\, Baker Building\, Division F meeting room
 \, 5th floor
END:VEVENT
END:VCALENDAR