An intelligent virtual assistant is provided for respectively customizable interactive audio/video content to each of a plurality of computing devices during a networked communication session. Input is received from at least one device, and is determined via information provided in or with the input, that the input is at least one of speech input, facial input, gesture input and textual input. The input is processed and machine-based learning occurs using at least some of the processed input. The intelligent virtual assistant is provided for the respectively customizable interactive audio/video content as a function of the machine-based learning process via at least one of the plurality of computing devices.