In order to design an English speech emotion recognition model with excellent classification performance,the hidden layer structure of the back-propagation algorithm is adjusted in the research,and the English speech emotion recognition model is composed of the combination of the back-propagation algorithm and the seagull optimization algorithm.The test results are as follows:The designed model has a loss function of 1.33 after training,which is lower than the comparison model.Its recognition accuracy and accuracy are 96.1％and 96.5％,respectively,which are also higher than all the comparison models.The test results prove that the English speech emotion recognition model designed in this study has excellent English speech recognition ability,and has certain ap-plication potential in improving the translation ability of English machine translation systems.
维普
万方数据