Speech Recognition Based on Genetic Algorithm Optimized Support Vector Machine

Abstract

With the development of artificial intelligence, intelligence is becoming more and more common. Among them, speech emotion recognition is an outstanding direction of intelligence and an important research direction of artificial interaction. Aiming at the problem of speech emotion recognition, this paper uses genetic algorithm to optimize support vector machine to recognize speech emotion. In the experiment, the Chinese academy of sciences language library was used for training and testing, and a wavelet packet based principal component analysis was used for feature extraction, and compared with the traditional speech emotion recognition. The experimental results show that the optimal support vector machine recognition rate is 95%.