Model Innovation and Practice of English Education in Colleges and Universities in the Context of Internet

In this paper, we use the particle filtering algorithm to create a tracking model. We then choose the target of the current frame as a candidate model and describe it using edge features and pixel ratio features. We then combine the extracted features to find the similarity between the target template and the candidate template and use the Bhattacharyya distance to show it. The hand position information obtained by Kinect is utilized to control the virtual hand. Correct the angle of the virtual hand to achieve a real-time correction of its position. According to the course content, define the user operation, analyze the intention of their interaction behavior, and design the inquiry-based interaction classroom. Recognize teacher and student behaviors using deep RNNs, with speech processing as the core, and combine with ResNet technology to detect the teaching scene in the classroom. Using the model of this paper in English education in colleges and universities, the teachers of the two teaching modes are analyzed for their teaching tendency, and for students’ speech, the two types of classrooms show obvious differences, with a p-value of 0.036 < 0.05. In the interactive English classroom teaching, the average score of the posttest of the experimental group is 12.786, which is 1.421 higher than the average score of the control group, and the two-tailed value of the significance is 0.041, which is less than 0.05. There is a significant difference between the two groups, and the interactive English teaching mode helps to improve students’ English listening performance.