This paper takes human-computer pragmatics as the theoretical core and integrates the optimization idea of "concretization of abstract knowledge" in Geometer's Sketchpad to construct a three-dimensional nine-point pragmatic optimization path of "context perception-interaction logic-personalized experience". Based on practical cases from automotive enterprises, this paper deeply analyzes the practical logic and application effects of core mechanisms such as dynamic context modeling and scenario-based corpus construction, providing support for automotive enterprises to
create natural and smooth English voice interaction systems that adapt to cross-cultural communication needs. This helps intelligent connected vehicles solve interaction problems in their global layout and further enhances the trust and usage experience of cross-regional users.