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

摘要

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 memoryrecurrent 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 compared with the manual recognition approach; the LSTM-RNN-based machine translation algorithm had the lowest error rate for the translation of speech recognition results, and the translation gained the highest rating in the evaluation of ten professional translators.