A real-time receiver and method for receiving and playing out real-time packetized data are disclosed. The receiver includes a packet transmission fixed delay estimator and a packet transmission variable delay estimator. The fixed delay estimator determines, using packets received up to the current point in a conference, the non-variable portion of observed delays. This non-variable portion is subtracted from each packet's observed delay to obtain a variable delay estimate for that packet.
Since variable delays actually drive the buffering time needed at the receiver to achieve smooth playout, the packet variable delay estimates can be used directly to adjust playout delay. Adaptive playout delay is preferably set aggressively low, based on observed packet variable delay estimates, to reduce data latency. Playout delay can be adjusted rapidly upwards when higher packet delays are observed, allowing rapid adaptation to network statistical variations and reducing the frequency of late packets.