Innovative Application of Digital Audiovisual Language Based on Natural Language Processing Technology in Graffiti Art Education

In this paper, natural language processing technology is used to establish coupled Hidden Markov Models for speech formalization and features of the two channels. The adaptive weight selection method is used to recognize speech under different signal-to-noise ratios in the recognition stage. The audio-visual recognition system is applied to actual graffiti art education research, and the feasibility and development direction of audio-visual graffiti in the classroom are studied through questionnaires. It can be found that most teachers and students believe that the concept of audiovisual graffiti painting is helpful or more helpful in the cultivation and development of students’ subconscious mind. Most teachers and students believed audiovisual doodling was beneficial or advantageous for developing students’ subconscious. In the acceptance test and learning effect analysis of the audiovisual graffiti course, the mean values of the four dimensions of perceived usefulness (41.87), perceived ease of use (18.11), willingness to use (31.45), and perceived experience of the process (28.31) are basically around the maximum value, which indicates that the audiovisual technology is well accepted and that the mean value of the experimental group’s performance score (83.68) is significantly higher than that of the control group. Control group. The audiovisual course has been found to enable learners to concentrate on the learning situation faster and master the teaching essential contents more thoroughly.