The development of a dance-musification model with the use of machine learning techniques under COVID-19 restrictions
Konstantinos Bakogiannis, Areti Andreopoulou, Anastasia Georgaki · 2021
Interactive technologies enable dancers to control the music in real-time with their movement. This paper presents the design and development of a model which takes as input a dancer’s movement and outputs music, structurally related to dance, with the use of machine learning techniques. Both the technical and artistic aspects of the model development are described in detail. In particular, the paper compares the use of machine learning techniques to traditional coding, in interactive dance and music applications. Moreover, it describes the significant discrimination between movement sonification and dance musification and explains why the model presented here falls into the second category. Special focus is given to the implications of the COVID-19 restrictions regarding the established collaboration with the dancer.