Siamese Residual Neural Network for Musical Shape Evaluation in Piano Performance Assessment

Understanding and identifying musical shape plays an important role in musiceducation and performance assessment. To simplify the otherwise time- andcost-intensive musical shape evaluation, in this paper we explore howartificial intelligence (AI) driven models can be applied. Considering musicalshape evaluation as a classification problem, a light-weight Siamese residualneural network (S-ResNN) is proposed to automatically identify musical shapes.To assess the proposed approach in the context of piano musical shapeevaluation, we have generated a new dataset, containing 4116 music piecesderived by 147 piano preparatory exercises and performed in 28 categories ofmusical shapes. The experimental results show that the S-ResNN significantlyoutperforms a number of benchmark methods in terms of the precision, recall andF1 score.