An apparatus, method and computer program is described comprising: receiving a plurality of the interaction inputs for one or more of a plurality of time slices of a virtual meeting or presentation from one or more participants of the virtual meeting or presentation; determining one or more significant time slices of the virtual meeting or presentation based, at least partially, on one or more of the plurality of interaction inputs; and generating a summary of the virtual meeting or presentation, wherein the summary comprises audio data corresponding to at least some of the one or more significant time slices.