In the last decades there have been several attempts to use computers in Music Education. New pedagogical trends encourage incorporating technology tools in the process of learning music. Between them, those systems based on Artificial Intelligence are the most promising ones, as they can derive new information from the inputs and visualize them in several meaningful ways. This paper presents an application of machine learning to music performance which is able to discover the similarities and differences between a given performance and those from other musicians. Such a system would help students to better learn how to perform a certain piece of music, allowing them to compare with other students or master performers. [For the full proceedings, see ED562127.]