| 基于神经网络和卡尔曼滤波算法的说话人识别 |
| |
| 引用本文： | 张余生,夏秀渝,杨莎.基于神经网络和卡尔曼滤波算法的说话人识别[J].成都信息工程学院学报,2008,23(4). |
| |
| 作者姓名： | 张余生 夏秀渝 杨莎 |
| |
| 作者单位： | 四川大学电子信息学院,四川,成都,610064 |
| |
| 摘 要： | 首先从语音信号中提取出特征参数:线性预测倒谱系数(LPCC)和用小波包提取的小波特征参数(WPC);语音特征分类模型则选择多层前馈式神经网络(MBP网络),并将奇异值分解运用到扩展卡尔曼滤波(EKF)算法中作为神经网络的学习算法.仿真结果表明,小波特征参数具有良好的识别效果;同时采用改进后的扩展卡尔曼滤波(EKF)算法使人工神经网络具有更稳定、更准确的分类性能.
|
| 关 键 词： | 线性预测倒谱系数(LPCC) 小波特征参数 多层前馈式神经网络 扩展卡尔曼滤波(EKF)算法 |
Speaker recognition based on artificial neuron network and Kalman filter algorithm |
| |
| ZHANG Yu-sheng,XIA Xiu-yu,YANG Sha.Speaker recognition based on artificial neuron network and Kalman filter algorithm[J].Journal of Chengdu University of Information Technology,2008,23(4). |
| |
| Authors: | ZHANG Yu-sheng XIA Xiu-yu YANG Sha |
| |
| Institution: | ZHANG Yu-sheng,XIA Xiu-yu,YANG Sha (College of Electronic Information,SCU,Chengdu 610064,China) |
| |
| Abstract: | The LPCC coefficients and the wavelet packet coefficients are distilled from the speech.They are combined with the multi-BP network so that Kalman filter learning algorithms make the text-independent recognition.The experiment results show that the wavelet packet coefficients have good performance and the Kalman filter learning algorithms improves the capability of the neural network. |
| |
| Keywords: | LPCC wavelet coefficient artificial neuron network Kalman filter algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
