<p>With the widespread application of wearable technology, the field of music education has ushered in new opportunities for personalized recommendations. However, the traditional music teaching model has shortcomings such as low level of informatization and single teaching content. This study aims to utilize wearable sensor technology and soft computing algorithms to explore how to achieve personalized recommendations for music information technology teaching, thereby improving teaching effectiveness and user experience. Research on selecting suitable wearable sensors to obtain physiological indicators and exercise data of students during music learning, monitor student information in real-time, and utilize soft computing algorithms for processing. By analyzing and mining data, reveal patterns such as emotional changes and fluctuations in focus among students during the learning process. Finally, by integrating sensor data and the results of soft computing algorithms, a corresponding algorithm model is designed, taking into account the personality characteristics, learning preferences, and characteristics of music content of students, in order to achieve personalized recommendations. By continuously optimizing model parameters and algorithm logic, the accuracy and personalization of recommendations are improved. The experiment verified the effectiveness of personalized recommendation based on wearable perception and soft computing algorithms in music teaching, which can better meet the personalized needs of students, improve teaching effectiveness and learning motivation.</p>