Application of Short Video Description Technology in College English Teaching

Introduction

For any college English teacher, it is of great significance to deeply explore the application of new teaching modes in college English teaching. Different from traditional teaching forms, micro-courses mainly refer to courses in which the teachers use a video recording to highlight key knowledge or difficult knowledge according to the specific requirements of practical guidance or teaching standards (Li, 2021). However, in the current context of junior high school English teaching, an important teaching error lies in the fact that some teachers focus too heavily on improving student achievement and ignore the important communicative function of English as a language subject, which greatly suppresses students' interest in learning and seriously affects the English teaching effect (Li, 2019). In order to avoid students’ negative attitude toward learning, English language teachers must abandon traditional teaching methods, adopt the use of micro-courses, focus on connecting with students, and create an efficient and unique teaching space to meet students’ learning needs. Therefore, finding a new teaching mode and structure through exploration and practice to expand students' exposure to English, create the best English learning situation, and stimulate students' interest in learning has become one of the key factors for students to achieve good teaching results.

Short video technology organically integrates language learning and cultural background knowledge learning. It organically combines phonetic learning with appreciating plots and pictures, which can effectively improve students' interest in learning English. As a language discipline, the learning goal of English should be for students to use it as a communication tool and to achieve the comprehensive development of listening, speaking, reading, and writing. In order to improve teaching efficiency and obtain higher test results, teachers in traditional classrooms usually transfer knowledge to students in one way. Students rarely have the opportunity to interact with teachers in language, and it is difficult to create an effective language learning environment. Applying short videos to English teaching can create an active and interesting classroom atmosphere. The pictures, sounds, scenes, and other aspects in the short video will have a sensory impact on students, attract students' attention, and enhance students' enthusiasm for learning English, achieving the effect of teaching and entertainment. Students can participate in the process of making short videos. The process of making short videos can improve pronunciation, improve their own accent standards, and enhance their confidence in speaking English through imitation of standard English accents.

A video consists of a group of shots, and the reprocessing of this video clip usually cuts these shots, so it is important to explore the shot structure of videos through video content analysis. As a language subject, English cannot be interpreted perfectly if it is taught only through traditional classroom methods (Li et al., 2018; Bera et al., 2021). The short video teaching can build a good language environment for students, encourage them to feel the connotation of English in real scenes, and strengthen their own learning by imitating English accents, which will help students strengthen their oral expression abilities. Moreover, even if there is oral English teaching, only some simple communicative situations are set up, and the teaching content is limited to exercises in textbooks, which limits the space for students' oral English development, hinders students' development, and further affects students' interest in English learning. This paper begins with an overview of the main characteristics of short video technology, then analyzes the value of short video technology in college English teaching, and designs a new generation of short video natural language model based on attention mechanism. Various modal features are input into an independent encoder for calculation, which can better learn the underlying semantic information contained in various features.