A computer-based narration assembly system and method provide an environment analogous to in-person sharing of photographs by utilizing spontaneous verbal narration or story-telling, together with manual indications by the story-teller of significant or relevant image portions. The system combines the verbal narration with automatic cinematic display manipulations that relate to the manual indications by the story-teller to form a multimedia production or "movie" from the display images. The cinematic display manipulations may include pans, zooms, fades, etc. that animate the display of the images and transitions between them.