基于决策树和改进SVM混合模型的语音情感识别

为有效提高语音情感识别的准确性，达到人机和谐交互的目的，本文提出了一种基于决策树和改进SVM混合模型的语音情感识别方法，有效地避免了无界泛化误差、分类器数目多、受限优化等问题，提高了悲伤、喜悦、愤怒、厌恶、惊讶、恐惧6种基本情感识别效率。实验结果表明，该方法识别准确率为87.58%，与传统的支持向量机和人工神经网络方法相比，有更高的抗噪声能力和稳定性，能得到更高的识别准确率，而且有较强的实用性和推广能力。 

Speech Emotion Recognition Based on Decision Tree and Improved SVM Mixed Model

To effectively improve the accuracy of speech emotion recognition in intelligent man-machine harmonious interaction, a method of speech emotion recognition was proposed based on decision tree and an improved SVM mixed model. This method can avoid the tree unbounded generalization error, more the number of classifiers and other shortcomings, while taking advantage of SVM-KNN mixed model to avoid constrained optimization problems and improve the recognition efficiency. In this paper, six basic emotions were identified, including sadness, joy, anger, disgust, surprise, fear. Experimental results show that this method can effectively identify six basic emotions. Compared with the traditional support vector machine and artificial neural network method, this method can get higher recognition accuracy, better stability, strong practicability and generalization ability.