A computerized method is provided that enables an interactive multimedia session between a group of geographically distributed musicians. The method includes song arrangements for the interactive multimedia session being specified as a sequence of song parts to be played or sung by each of the participating geographically distributed musicians. Each musician performance is automatically detected on an instrument track along with audio and video for each musician performance on any song part. The timing for each musician performance is automatically captured by the system. The captured performances are transmitted to the musicians participating in a same session of the geographically distributed musicians to produce the effect of playing with other musicians live in the interactive multimedia session. A computer-implemented system and a computer program product stored on a non-transitory computer-readable storage medium for practice of the method are also provided.