Generating Different Story Tellings from Semantic Representations of Narrative
This addresses the need for interactive story systems to produce varied narratives for different audiences, though it appears incremental as it builds on existing semantic representations and NLG engines.
The paper tackled the problem of automatically generating different story tellings from semantic representations of narrative by developing a method to convert Scheherazade's story intention graph into input for the Personage NLG engine, achieving results where generated stories were close to a customized realizer as measured by Levenshtein Distance and BLEU scores on 35 Aesop Fables.
In order to tell stories in different voices for different audiences, interactive story systems require: (1) a semantic representation of story structure, and (2) the ability to automatically generate story and dialogue from this semantic representation using some form of Natural Language Generation (NLG). However, there has been limited research on methods for linking story structures to narrative descriptions of scenes and story events. In this paper we present an automatic method for converting from Scheherazade's story intention graph, a semantic representation, to the input required by the Personage NLG engine. Using 36 Aesop Fables distributed in DramaBank, a collection of story encodings, we train translation rules on one story and then test these rules by generating text for the remaining 35. The results are measured in terms of the string similarity metrics Levenshtein Distance and BLEU score. The results show that we can generate the 35 stories with correct content: the test set stories on average are close to the output of the Scheherazade realizer, which was customized to this semantic representation. We provide some examples of story variations generated by personage. In future work, we will experiment with measuring the quality of the same stories generated in different voices, and with techniques for making storytelling interactive.