Exploring Vocal Learning Practices From a Deep Learning Perspective: A Qualitative Case Study

Abstract

Vocal instruction in higher music education requires more than technical mastery; it also involves students’ awareness of their vocal processes, understanding of the relationship between vocal technique and musicality, reflective abilities, strategic exploration, and the capacity to transfer acquired skills across different contexts. This study aims to explore vocal learning practices from a deep learning perspective and identify pedagogical needs as a foundation for developing an innovative deep learning model for vocal instruction. The study employed a qualitative approach using an exploratory case study design. Data were collected through observations of vocal learning sessions, semi-structured interviews with vocal instructors and students, and analysis of relevant learning documents and artifacts. The data were analyzed using reflexive thematic analysis through the processes of data familiarization, initial coding, theme development, theme review, theme definition, and interpretation. The findings identified six main themes: (1) vocal instruction is predominantly oriented toward technical mastery and instructor-led correction; (2) students’ vocal awareness is still developing; (3) vocal technique and musical meaning are not always integrated; (4) student engagement has not fully developed into learner autonomy; (5) reflection has not yet become a systematic component of the learning cycle; and (6) the transfer of vocal skills across different contexts remains limited. These findings indicate that the development of deep vocal learning should strengthen vocal awareness, the integration of technique and musicality, exploration, student agency, reflection, and learning transfer. These six aspects provide an empirical foundation  for formulating the characteristics and design requirements of an Innovative Deep Learning Model for Vocal Instruction.