Design And Implementation Of An Interactive Course System

In recent years,with the improvement of the status of Chinese,our country has paid more and more attention to the quality education of Chinese.In the compulsory education stage,the proportion of ancient poetry and prose is increasing year by year.However,the reading comprehension ability of students in the primary school stage is generally poor,and students face a series of problems such as difficulty in selecting books,poor reading concentration,poor reading effect,and lack of reading interest.In addition,reading is highly non-standardized,students are highly dependent on teachers,and high-quality teachers who stimulate students’ interest in reading are scarce.With the implementation of the national education policy on reducing the burden on students,schools also need to use their spare time scientifically to carry out appropriate reading and literary activities.Therefore,efficient reading has become a rigid demand in the market.In order to meet the above requirements,this paper intends to develop an interactive classroom system based on speech recognition.Starting from the two roles of course producer and course consumer,the system designs and implements functions such as authority management module,course generation module,real-time barrage,group confrontation,and game interaction.Among them,The authority management module includes functions such as login and registration,personal center,and authority verification.The course generation module builds a set of standard interactive classroom generation process.By modularizing the course content and standardizing the editing process,the efficiency of course development is improved.At the same time,multiple common components are extracted,which improves the reusability and maintainability of the system.The course generation module builds a standard interactive classroom generation process.By modularizing the course content and standardizing the editing process,the efficiency of course development is improved.Multiple common components are extracted at the same time,which improves the reusability and maintainability of the system.The interactive classroom creatively introduces speech recognition into the primary school Chinese teaching scene,realizing a variety of interactive scenes with strong interactive experience.The interactive scene includes the real-time barrage function of answering questions to generate barrage,the interesting group confrontation function,and the game interaction function based on voice.For the speech recognition technology in the interactive classroom system,this paper introduces the processing flow of Kaldi and speech recognition in detail.It focuses on the weighted finite state transformer(WFST)under Kaldi,analyzes the realization of WFST and the method of constructing decoding graph based on WFST,and gives the method of constructing WFST online.And builds the GMM-HMM and DNN-HMM models through Kaldi.By comparing the word error rate,and training time,the advantages and disadvantages of the two models are expounded.The interactive classroom system is developed in a way of separating the front and back ends.The front-end is developed using mainstream front-end methods such as React framework and Pixi.js.The server uses Java as the development language,uses the SSM framework,and uses the My SQL database for data storage.The front and back ends interact through AJAX,and the interface is uniformly designed in a RESTful style.