Interaction with multi-modal picture through image feature selection and voice query

A system for presenting a multi-modal picture includes picture presentation equipment (37) for displaying an image (38) of the picture and for enabling a user to interact with the picture by selecting specific picture features and asking a specific query relating to the feature. A voice browser system (3) controlled according to dialog scripts (46) associated with the picture, determines an appropriate response having regard to the spoken user query and the picture feature being pointed to. Each picture can have multiple narrators associated with it and the can choose which narrator is currently active. Picture authoring apparatus is also provided. The image used to specify a particular feature may be a hard copy.