Research on the Efficiency of Intelligent Algorithm for English Speech Recognition and Sentence Translatione

Abstract

Machine translation has been gradually widely used to improve the efficiency of English translation. This paper briefly introduced the English speech recognition algorithm based on the back-propagation (BP) neural network algorithm and the machine translation algorithm based on the long short-term memory-recurrent neural network (LSTM-RNN) algorithm. Then, the machine translation algorithm was simulated and compared with BP-RNN and RNN-RNN algorithms. The results showed that the BP neural network algorithm had a lower word error rate and shorter recognition time than manual recognition; the LSTM-RNN-based machine translation algorithm had the lowest error rate of the translated words in the speech recognition results, and the translation gained the highest evaluation score from ten professional translators.