Interpretable and multimodal framework for real-time affective feedback in music practice


Abstract

An affect-aware, AI-driven feedback system tailored for music education—one that meaningfully integrates real-time emotion recognition with instructionally sound intervention strategies. Traditional teaching approaches, by and large, treat performance as a purely technical endeavor, often neglecting the fluctuating emotional and cognitive states that profoundly shape how students learn, express themselves, and persist through challenges. As a result, feedback tends to be generic, delayed, and ill-equipped to address the affective barriers—such as anxiety, frustration, or disengagement—that can impede musical development. To address this gap, we present a multimodal framework that jointly analyzes physiological signals and behavioral indicators to infer learners’ emotional states during practice. Trained on a curated dataset of 100 expert-annotated practice sessions, our FT-Transformer classifier distinguishes eight emotion categories grounded in psychophysiological taxonomy with 95\% accuracy and a macro-F1 score of 0.75—a balance of precision and recall that reflects both class diversity and real-world variability. Importantly, the model’s decisions are not opaque. This alignment not only validates our feature engineering but also reinforces the physiological plausibility of the system’s inferences. In essence, this work demonstrates that effective AI in education need not choose between technical sophistication and pedagogical relevance. By tightly coupling interpretable affect recognition with discipline-specific instructional logic, we offer a transparent, scalable blueprint for intelligent tutoring systems that respond not just to what a student plays, but to how they feel while playing. Given its modular design and grounding in universal affective mechanisms, the framework shows strong potential for adaptation beyond music—particularly in other high-stakes, affect-sensitive domains such as language acquisition, public speaking, or even therapeutic skill training.
Ask to review this manuscript

Notes for potential reviewers

  • Volunteering is not a guarantee that you will be asked to review. There are many reasons: reviewers must be qualified, there should be no conflicts of interest, a minimum of two reviewers have already accepted an invitation, etc.
  • This is NOT OPEN peer review. The review is single-blind, and all recommendations are sent privately to the Academic Editor handling the manuscript. All reviews are published and reviewers can choose to sign their reviews.
  • What happens after volunteering? It may be a few days before you receive an invitation to review with further instructions. You will need to accept the invitation to then become an official referee for the manuscript. If you do not receive an invitation it is for one of many possible reasons as noted above.

  • PeerJ Computer Science does not judge submissions based on subjective measures such as novelty, impact or degree of advance. Effectively, reviewers are asked to comment on whether or not the submission is scientifically and technically sound and therefore deserves to join the scientific literature. Our Peer Review criteria can be found on the "Editorial Criteria" page - reviewers are specifically asked to comment on 3 broad areas: "Basic Reporting", "Experimental Design" and "Validity of the Findings".
  • Reviewers are expected to comment in a timely, professional, and constructive manner.
  • Until the article is published, reviewers must regard all information relating to the submission as strictly confidential.
  • When submitting a review, reviewers are given the option to "sign" their review (i.e. to associate their name with their comments). Otherwise, all review comments remain anonymous.
  • All reviews of published articles are published. This includes manuscript files, peer review comments, author rebuttals and revised materials.
  • Each time a decision is made by the Academic Editor, each reviewer will receive a copy of the Decision Letter (which will include the comments of all reviewers).

If you have any questions about submitting your review, please email us at [email protected].