In order to improve the innovation effect of English teaching mode, this paper applies Augmented Reality (AR) and Virtual Reality (VR) technology to the innovation of college English teaching mode to promote the improvement of English teaching quality. Moreover, this paper accelerates ray tracing through Compute Unified Device Architecture (CUDA)’s parallel operations, hierarchical bounding box space structure, and a series of optimization methods for Monte Carlo estimation sampling convergence. Furthermore, this paper computes global illumination by collecting lighting information in textures through voxel cone tracking. In addition, this paper proposes attenuation based on normal weights, which can reduce the normal consistency problem of voxel direct lighting calculations. Through the experimental research, it can be seen that the college English teaching model based on VR/AR technology proposed in this paper can effectively improve the effect of English teaching