论文部分内容阅读
提出一种基于径向基函数神经网络的改进聚类方法,并将此改进的神经网络应用于语音识别领域,建立一个非特定人的孤立词语音识别系统.此聚类方法采取有监督的学习方式,将训练样本的形心作为隐节点的质心,训练样本的分类数作为隐节点的个数.利用该方法对小词表汉语孤立词进行语音识别.结果表明,采用此算法的径向基函数的神经网络具有更好的分类能力,训练速度和识别率均优于传统的径向基函数网络.
This paper proposes an improved clustering method based on radial basis function neural network and applies this improved neural network in the field of speech recognition to establish a non-specific isolated speech recognition system.This clustering method adopts supervised learning The centroid of the training sample is taken as the centroid of the hidden node and the number of training samples as the number of hidden nodes.The speech recognition of Chinese isolated words in the small vocabulary is performed by using this method.The results show that the radial basis Function neural network has better classification ability, training speed and recognition rate are better than the traditional radial basis function network.