We use rhetorical annotations to specify a generation process that can assemble meaningful video sequences with a communicative goal and an argumentative progression. Our annotation schema encodes the verbal information contained in the audio channel, identifying the claims the interviewees make and the argumentation structures they use to make those claims. Based on this schema, we construct a semantic graph which is traversed by rhetoric-based strategies selecting video segments. The selected video segments are edited to form a meaningful video sequence.

One of the challenges of multimedia applications is to provide user-tailored access to information encoded in different media. Particularly, previous research has not yet fully explored how to automatically compose different video segments according to a communicative goal. We propose a rhetoric-based method to support the selection and automatic editing of user-requested content from video footage. The method is applied to the domain of video documentaries to create biased sequences about a user selected subject.