Narrative Variations in a Virtual Storyteller

TL;DR

This paper introduces Fabula Tales, a computational framework that generates varied storytelling styles by manipulating narratological parameters, impacting reader perception using deep story representations from diverse sources.

Contribution

It presents a novel, generalizable method for creating multiple story tellings from a single content source through a deep narrative representation.

Findings

Different tellings influence reader perception of stories and characters.

The framework effectively reuses existing story content for varied storytelling.

User studies confirm the impact of narratological variations.

Abstract

Research on storytelling over the last 100 years has distinguished at least two levels of narrative representation (1) story, or fabula; and (2) discourse, or sujhet. We use this distinction to create Fabula Tales, a computational framework for a virtual storyteller that can tell the same story in different ways through the implementation of general narratological variations, such as varying direct vs. indirect speech, character voice (style), point of view, and focalization. A strength of our computational framework is that it is based on very general methods for re-using existing story content, either from fables or from personal narratives collected from blogs. We first explain how a simple annotation tool allows naive annotators to easily create a deep representation of fabula called a story intention graph, and show how we use this representation to generate story tellings automatically. Then we present results of two studies testing our narratological parameters, and showing that different tellings affect the reader's perception of the story and characters.