Rating Practice Piano Pieces

Autonomous rating of music performance could better the performers' understanding of playing techniques and provide a flexible tool for the evaluation of autonomous music generation.  In this project, we adopt multiple features of the music signal processing models to automate the evaluation of a large set of piano pieces. The scoring of piano pieces is divided into different music dimensions, among which the pitch, the tempo and the timbre are the dominating factors. With the aid of this division, our project is capable of presenting individual scoring for different music dimensions and the integrated scoring for the overall performance. Apart from the large set of piano pieces played by different performers, personally-played piano pieces are also included in the test samples to validate the model.