Research And Implementation Of Music Intelligent Processing System For Scratch Platform

With the rapid development of computer technology,artificial intelligence technology has been applied in all walks of life.In the field of information science education,block-based programming languages have received more and more attention from students and teachers.Scratch is a block-based visual online programming language.It uses children’s building blocks to visually represent code modules.The low threshold and easy-to-use characteristics make it popular with more and more students and teachers.The music module makes Scratch projects more diversified and artistic.The music in projects can express the author’s emotions well.However,the current Scratch platform does not provide a tool that can recognize music emotions,and users cannot understand the emotions of Scratch music they produce.At the same time,the Scratch platform does not provide a music library for users to retrieve relevant music for their projects according to their needs.Therefore,the application of artificial intelligence in Scratch music processing system not only enables students to develop more vivid projects,but also intuitively experience and understands artificial intelligence technology.To design and implement a music intelligent processing system for the Scratch platform,it is necessary to conduct research on issues such as music emotion recognition and music generation,and build a Scratch music emotion recognition model and a Scratch music generation model.In emotion recognition model,the Scratch music emotion recognition data set is constructed by analyzing the differences between Scratch and traditional music.Then the Scratch music’s low-level audio characteristics and timing characteristics are used to learn the relationship between Scratch music and emotional concepts.Moreover,the attention mechanism is introduced to improve the recognition rate.Through generative adversarial networks based on emotion and time structure,a Scratch music generation model is constructed to generate the Pianoroll format music.Moreover,the convolutional neural network is used in the generator and discriminator to improve the training speed.The two algorithm models have verified the effectiveness and practicability of the designed models through comparative experiments.Finally,a music intelligent processing system is designed and implemented for the Scratch platform.By integrating the music emotion recognition model and the music generation model into the system,Scratch music emotion recognition,Scratch music retrieval and other related functions are realized.The system enriches the music-related modules in the Scratch platform and improves the fun of the platform.Additionally,it helps users to create projects and learn programming better.The young people can have a personal experience of artificial intelligence when using the music-related functions of the Scratch platform,which will increase the interest of young people in learning programming.