Automatic Music Error Detection

Researchers at Purdue University have developed an end-to-end, transformer-based trainable model that automatically detects musical errors. Their model, called Muse, helps beginner musicians identify specific errors in their performances, such as playing incorrect notes or rhythms, and subsequently provides automatic feedback to learners and their instructors. Existing tools for music error detection face a number of limitations, including reliance on error-prone automatic alignment and a lack of sufficient data on which to train, resulting in over-reliance on heuristics. Muse overcomes these issues through its novel transformer-based design that can be trained to implicitly align and compare performance audio with music scores through latent space representations. Moreover, Muse is compatible with multiple instruments and aligns users' performance with a provided reference.

Technology Validation:

Muse was tested against a baseline, representing existing similar music error detection models, for performance and accuracy. Muse demonstrated superior accuracy detecting most error types, except for a higher recall for extra notes in the baseline model. Muse also showed consistently higher F1 Error Detection scores than the baseline across all categories.

Advantages:

-Adaptable for use with multiple instruments

-Automatically aligns users' performances with audio reference

-More sophisticated error detection, including missed, extra, or incorrect notes

-Greater accuracy than similar models

Applications:

-Music education, including self-teaching and assistance to instructor-based learning

Publications:

Chou, Ben S.-H., et al. "Detecting Music Performance Errors with Transformers." School of Electrical and Computer Engineering, Purdue University, 2025. Available at: https://arxiv.org/pdf/2501.02030

TRL: 4

Intellectual Property:

Keywords: AI music error detection,Transformer-based music analysis,Automatic music transcription,Performance feedback tools,Music education technology,Multi-instrument compatibility,Audio-score alignment,Music learning assistance,Intelligent music tutoring,Music performance analytics,Error detection algorithms,Adaptive music software,Digital music training tools,Music practice evaluation