Beyond output metrics: reframing AI-assisted vocal pedagogy through human learning and educational value

Artificial intelligence is increasingly entering music and performance education, often through tools that measure pitch, stability, timing, and related performance indicators. In vocal pedagogy, however, more precise measurement does not automatically create educational value. This Perspective argues that AI-assisted vocal pedagogy should be evaluated by how AI-generated evidence becomes meaningful for human learning: how learners interpret feedback, regulate practice, sustain motivation, and develop trust in teacher-guided learning processes. The conceptual problem is that measurable vocal outputs may not fully represent learners' internal experience, embodied coordination, or expert pedagogical interpretation. Drawing on educational psychology, performance science, and recent work on AI-supported and metacognitive music learning, the article proposes a framework that links technical adaptation, human learning processes, and educational outcomes. Within this framework, effectiveness, equity, and sustainability are treated as outcome criteria for judging whether AI-supported feedback becomes meaningful for learning, usable across learners, and sustainable for long-term vocal development. The article concludes that AI should not be positioned as an autonomous evaluator of singing quality, but as a human-centered support for interpretation, reflection, teacher-student dialogue, and pedagogically responsible decision-making.