This study reports the design and preliminary evaluation of a smart classroom environment for undergraduate music majors. A one-semester quasi-experimental pretest–posttest design was implemented at a university in Xi’an, China, in two intact classes: an experimental group receiving smart-classroom instruction (n = 36) and a control group receiving conventional instruction (n = 36). The intervention embedded digital music resources, AI-assisted creative tools, platform-based learning activities, and data-informed feedback into regular course teaching. Data were collected from student questionnaires, platform logs, teaching records, and course-product evaluations. Seven operational indicators were compared descriptively: digital resource support, participation in smart-classroom activities, teacher–student interaction timeliness, evaluation quality, learning autonomy, student output quality, and exploratory instructional cost-efficiency. Group-level results indicated improvement in both groups, with larger descriptive gains in the experimental group. The digital resource richness index increased from 42 to 81 in the experimental group and from 43 to 65 in the control group; the cost-efficiency ratio increased from 0.60 to 0.82 and from 0.61 to 0.69, respectively. Because individual-level data were unavailable, standard deviations and inferential tests were not reported. The findings offer preliminary descriptive evidence that smart-classroom design may support resource access, participation, feedback, and self-directed learning in music-major teaching.