The development of intelligible oral pronunciation in deaf and hard-of-hearing preschool children remains one of the most complex and debated issues in special pedagogy. Traditional articulation-based and tactile-mechanical methods often lead to artificial speech patterns that differ significantly from natural spoken language. This article presents a scientific analysis of a bisensory (auditory–visual) approach to teaching pronunciation, grounded in systematic auditory training and the use of sound-amplifying technology. Drawing on long-term experimental observations, the study examines the principles, stages, and outcomes of pronunciation development under conditions approximating natural language acquisition. The findings demonstrate that bisensory perception enables spontaneous phonetic generalization, natural rhythm and intonation, and individualized trajectories of speech development. The results support the effectiveness of the bisensory approach as a foundational model for early pronunciation training in deaf and hard-of-hearing children. Keywords: bisensory approach, pronunciation development, deaf preschoolers, auditory–visual perception, speech rehabilitation, phonetic rhythmics