Communications of the ACM
AI System Reads Novels, Writes Music for Them
Credit: iStockPhoto.com
Researchers have trained an artificial intelligence system to read works of fiction and create music based on the texts. The TransProse project is a collaboration between Hannah Davis of New York University, who explored the idea for a thesis project, and Saif Mohammad, a researcher at the National Research Council Canada, who created a word-emotion lexicon.
The system first reads the text of a work, then based on word scans, it assigns densities of two different states, positive or negative, and eight different emotions — joy, sadness, anger, disgust, anticipation, surprise, trust, and fear. The system then composes a musical piece that chronologically follows the novel, broken up into beginning, early middle, late middle, and end parts. The tempo, key notes, octaves, and other musical variables are determined by the emotional density data. The system renders Joseph Conrad's "Heart of Darkness" as a brooding progression in C minor with fear and sadness as its themes.
"One practical application might be in online book stores where a customer can click on a button to listen to the emotional tone of a book before deciding to buy it," Mohammad says.
From IDG News Service
View Full Article
Abstracts Copyright © 2014 Information Inc., Bethesda, Maryland, USA
No entries found