Speech Recognition Research Based On Probabilistic Neural Network of RBF

摘要：

Automatic speech recognition(ASR) is an important topic to be performed by a computer system. This paper improved the radial basis function(RBF) nerual network,and put forward the radial basis function probabilistic neural network(RBFPNN),which is a combination of RBF and Probabilistic Neural Network (PNN). All of the classifiers use a hybrid Linear Prediction Cepstrum Coefficient(LPCC) and MelFrequency Cepstrum Coefficient(MFCC) as their features for classification. RBFPNN is used for pattern classification and recognition,which improved flexibility and recognition performance. A recognition accurary of 89.7％ could be achieved from this experiment. The experiment shows that BRFPNN improves the recognition accurary and speed.

关键词： RBF neural network PNN Radial basis function probabilistic neural network ASR

作者: Li zhongqi Qin fang Yu shuiyuan Shang wenqian

作者单位: Communication University of China,Department of Computer Science,Beijing,China Communication University of China,Department of Liberal Arts ,Beijing,China

会议类型: 国际会议

会议名称: 2011 International Conference on Opto-Electronics Engineering and Information Science(2011光电电子工程与信息科学国际会议 ICOEIS 2011)

会议地点: 西安

会议语种:英文

页码: 1542-1545

在线出版日期: 2011-12-23（万方平台首次上网日期，不代表论文的发表时间）