In this work, the problem of teaching English in colleges and universities throughout the information age is addressed by using augmented reality (AR) technology to finish designing the teaching activities and objectives. It establishes the English teaching evaluation index system, uses the Hummingbird optimization algorithm to optimize the convolutional neural network’s parameters, and builds the English teaching quality evaluation model based on this optimization. We assess students’ happiness and their capacity for autonomous learning in English instruction in the information era by statistical analysis, and we investigate novel approaches for English teaching tutorials in the information age. The results show that, in terms of overall satisfaction with English teaching, freshmen students have the highest level of satisfaction with the three phases of English teaching, with values of 216.71, 215.36, and 218.13, and sophomores have the lowest level of satisfaction, with values of 137.41, 133.01, and 138.76, indicating that there is a difference in student satisfaction by grade level. Establishing goals and monitoring the use of strategies were highly correlated with teacher support (P=0.534, 0.554 >0.5), which indicates a significant correlation between teacher support and students’ ability to learn English independently. This study can have a positive impact on the creative impact of teaching English in colleges and universities and successfully promote innovative transformation of the teaching mode.