Application of Speech Recognition Translator based on Evolutionary Multi-objective Optimization Algorithm

By utilizing evolutionary algorithms to achieve intelligent control of English translation software and improve the accuracy and adaptability of translation, an English translation speech recognition and translation algorithm based on evolutionary multi-objective algorithms is proposed and applied. The improved evolutionary algorithm can significantly improve the accuracy of word recognition without increasing iterations. A multi-objective learning parameter model for speech recognition English translators has been established. A multi-objective optimization model for information processing was constructed through repeated iterations and probabilistic reasoning of speech sources. Through adaptive correction control method, single objective control is carried out under multi-objective optimization, and the output of speech recognition English translator is optimized based on multi-objective optimization constraints. The simulation results show that the parameter estimation accuracy of using this method for English translation software design and multi-objective routing optimization is high, with a maximum speech recognition accuracy of 97.13% and a maximum recognition time of only 20.314ms.