Rapid prototyping: using Wizard of Oz to emulate machine learning features for interactive artistic applications

Abstract

In this paper, we present using the Wizard of Oz method to rapid prototyping machine learning features in interactive artistic applications. Machine learning systems often require time and resources until they were available to be executed. But in the era of agile movement where we have to test as soon as possible and in artistic solutions whose scope initially perhaps be unclear, there is a need for a way to fast testing hypothesis. We also briefly described Ama, a system for adult ballet training at home, which served as a proof of concept of our strategy.