基于BP网络的汉语普通话声调识别

Mandarin Chinese tone recognition based on back-propagation neural network

研究了一种常用的模式分类器——BP神经网络,分析了BP网络的训练及识别过程,提取了能体现声调特性的特征数据组成分类特征向量,设计了具有一个隐含层的3层前馈网络作为分类器,对普通话声调样本库做了分类识别实验,分析了不同隐含层节点数的识别实验结果.实验结果表明,提取的音频特征基本有效,分类效果良好,具有一定的应用价值.

The BP neural network is studied and applied to tone recognition in this paper, which is a frequently used pattern classifier. Based on analysis of training and recognition processes of BP neural network, we extract feature data reflecting the tone characteristics to build the classification eigenvector. The classifier is designed as a 3- layer BP neural network with one hidden layer, and applied to recognize the four tones of Mandarin Chinese. The recognition results by BP neural network with different hidden layer nodes are compared. The experiment result indicates that the audio feature extracted by this classifier is valid for Mandarin Chinese, and verifies its good classification performance.