Detecting Music Performance Errors with Transformers

Abstract

Beginner musicians often struggle to identify specific errors in their performances, such as playing incorrect notes or rhythms. There are two limitations in existing tools for music error detection: (1) Existing approaches rely on automatic alignment; therefore, they are prone to errors caused by small deviations between alignment targets. (2) There is a lack of sufficient datasets to train music error detection models. In this work, we introduce a transformer-based approach that takes both audio and aligned score representations as inputs and learns to identify pitch and rhythm errors without depending on perfect alignment. We also release a synthesized dataset of erroneous performances built on top of the MusicNet corpus to enable training and evaluation of error detection models.

Authors

Venue

Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 39, Issue 22, pp. 23687–23695 · 2025