Teaching Behavior Analysis System Based On Voiceprint Recognition

With the comprehensive promotion of education informatization 2.0,information technology has been gradually integrated into all aspects of education and teaching,and now plays an unprecedented role in this area.It is the trend of education reform to apply information technology to teaching evaluation and to promote the development of education evaluation to intelligence and normalization.Traditional teaching evaluation methods mainly rely on manual work.This kind of methods have heavy workload,low efficiency,and more rely on the subjective judgment of evaluators,lacking of objectivity,and can not be applied normally.In the teaching reflection stage after class,teachers urgently need a kind of classroom analysis and evaluation tool which is efficient and intelligent,and can support real-time feedback and normalization analysis.Discourse is the main carrier of knowledge transmission in teaching activities.All of the dialogues between teachers and students constitute the process of knowledge transmission.Through analysing classroom discourse,we can get the teaching behavior,mode,and atmosphere of classroom,and provide auxiliary tools for teachers to reflect on their teaching behavior and improve their teaching capability.According to the characteristics of classroom discourse,this paper applies voiceprint recognition technology to the conventional S-T analysis method,designs and implements an intelligent teaching evaluation system,and displays the evaluation results visually.The main works of this paper are as follows.Firstly,based on the comparative study of several commonly used teaching behavior analysis methods,this paper chooses the traditional S-T analysis method as the basis,and import voiceprint recognition technology to build an intelligent classroom teaching behavior analysis system.Secondly,according to the sampling and recognition characteristics of voiceprint recognition algorithm,this paper makes some adaptive adjustments to the coding and index calculation mode of S-T analysis,and proposes an improved S-T analysis method based on voiceprint recognition,which can analyze and describe teaching behavior more accurately.Thirdly,this paper designs and develops a web-based classroom teaching behavior analysis system.The system can automatically classify and recognize the speaker’s identity in the video,and then encode the behavior of teachers and students,calculate the index of S-T analysis,judge the teaching mode of the class,and display the analysis results in the form of graphs.Finally,12 excellent courses on the platform of "one teacher one excellent course"are selected as test samples to test the function and performance of the teaching behavior analysis system.The effectiveness of the system is verified by comparing the results of intelligent analysis with the results of traditional manual analysis.Taking two of them as examples,this paper shows how to use the analysis results of this system for teaching reflection,so as to improve the follow-up teaching efficiency.