Application of Adaptive Genetic BP Model in Classification of Speech Characteristic Signal

Abstract

IA121 view of the traditional BP model in classification of speech characteristic signal, the model has low classification accuracy, poor stability, easy to fall into the local minimum, and so on, therefore, the adaptive genetic algorithm is introduced to the optimization of BP model parameters to overcome the problems of the model. The adaptive genetic algorithm has strong adaptability and robustness, so the genetic algorithm is used to optimize the weights and threshold parameters of the BP model, which can effectively improve the performance of the neural network. Establish the classification system of speech characteristic signal based on adaptive genetic neural network, using the folk song, guzheng, rock and pop four different music speech characteristic signal to train and test the classification model. The test results show that, compared with the traditional BP model, the classification model of speech characteristic signal based on adaptive genetic neural network has higher classification accuracy.