Sansa: A Modified Sansula for Extended Compositional Techniques Using Machine Learning by McLean J Macionis & Ajay Kapur
Sansa is an extended sansula, a hyper-instrument that is similar in design and functionality to a kalimba or thumb piano. At the heart of this interface is a series of sensors that are used to augment the tone and expand the performance capabilities of the instrument. The sensor data is further exploited using the machine learning program Wekinator, which gives users the ability to interact and perform with the instrument using several different modes of operation. In this way, Sansa is capable of both solo acoustic performances as well as complex productions that require interactions between multiple technological mediums. Sansa expands the current community of hyper-instruments by demonstrating the ways that hardware and software can extend an acoustic instrument's functionality and playability in a live performance or studio setting.