Application of big data language recognition technology and GPU parallel computing in English teaching visualization system

With the development of multimedia technology and network technology applications, it is possible to implement online teaching systems in schools. This article aims to realize the design of online English teaching system based on interactive speech recognition system. The teaching system uses the characteristics of English course learning to develop an online English teaching design system based on interactive speech recognition technology. This article uses a deep learning model in the process of establishing the model, using mathematical methods, and fitting a nonlinear function while performing linear calculations. In order to improve the recognition accuracy, the DNN-HMM model is used, which greatly improves the recognition rate of the online English teaching system. The need for labeling speech frames is more urgent in traditional training acoustic models, and many problems have arisen as a result. The step of labeling speech requires strong professionalism and a large workload, and there is no way to adapt to the processing of massive data. The system established in this paper no longer requires a lot of work for annotation, but combined with the CTC layer, the cyclic neural network plays an important role in the process of processing speech sequence signals, and the relationship between data can be used in the process to build LSTM-The CTC model makes data processing faster, and finally makes the English teaching system work more fluently and improves work efficiency, which is of great significance. In the training, this paper discovered the characteristics of LSTM network training such as large amount of calculation. Extract the command from the voice message, and then proceed further. Through the establishment of this system, it meets the needs of English teaching in most schools, enables students to communicate with teachers online, can better achieve learning goals and obtain good benefits.