Skip to main content Link Search Menu Expand Document (external link)

Chapter 12. Musical Robots: Overview and Methods for Evaluation

Author: Emma Frid

Abstract: Musical robots are complex systems that require the integration of several different functions to successfully operate. These processes range from sound analysis and music representation to mapping and modeling of musical expression. Recent advancements in Computational Creativity (CC) and Artificial Intelligence (AI) have added yet another level of complexity to these settings, with aspects of Human–AI Interaction (HAI) becoming increasingly important. The rise of intelligent music systems raises questions not only about the evaluation of Human-Robot Interaction (HRI) in robot musicianship but also about the quality of the generated musical output. The topic of evaluation has been extensively discussed and debated in the fields of Human–Computer Interaction (HCI) and New Interfaces for Musical Expression (NIME) throughout the years. However, interactions with robots often have a strong social or emotional component, and the experience of interacting with a robot is therefore somewhat different from that of interacting with other technologies. Since musical robots produce creative output, topics such as creative agency and what is meant by the term “success” when interacting with an intelligent music system should also be considered. The evaluation of musical robots thus expands beyond traditional evaluation concepts such as usability and user experience. To explore which evaluation methodologies might be appropriate for musical robots, this chapter first presents a brief introduction to the field of research dedicated to robotic musicianship, followed by an overview of evaluation methods used in the neighboring research fields of HCI, HRI, HAI, NIME, and CC. The chapter concludes with a review of evaluation methods used in robot musicianship literature and a discussion of prospects for future research.

PDF

Robots mentioned in the chapter: