| 基于遗传算法和小波神经网络的语音识别研究 |
| |
| 引用本文： | 斯芸芸,徐道连,周卓然.基于遗传算法和小波神经网络的语音识别研究[J].微型机与应用,2011,30(16):68-71. |
| |
| 作者姓名： | 斯芸芸 徐道连 周卓然 |
| |
| 作者单位： | 重庆大学光电工程学院,重庆,400030 |
| |
| 摘 要： | 小波神经网络算法(WNN)易陷入局部极小,收敛速度慢,全局搜索能力弱,而遗传算法(GA)具有高度并行、随机、自适应搜索性能和全局寻优的特点。因此,将遗传算法和小波神经网络结合起来形成一种训练神经网络的混合算法——GA-WNN算法。仿真实验结果表明,该算法有效地缩短了识别时间,提高了网络训练速度和语音的识别率。
|
| 关 键 词： | 语音识别 小波神经网络 遗传算法 GA-WNN |
Research of speech recognition based on genetic algorithm and wavelet neural network |
| |
| Si Yunyun,Xu Daolian,Zhou Zhuoran.Research of speech recognition based on genetic algorithm and wavelet neural network[J].Microcomputer & its Applications,2011,30(16):68-71. |
| |
| Authors: | Si Yunyun Xu Daolian Zhou Zhuoran |
| |
| Affiliation: | Si Yunyun,Xu Daolian,Zhou Zhuoran(College of OptoElectronic Engineering of Chongqing University,Chongqing 400030,China) |
| |
| Abstract: | The wavelet neural network algorithm （WNN）easily leads to failing into local minimum, slow convergence rate, and weak global search capability, while the genetic algorithm （GA）has the characteristics of height parallel performance, random, adaptive search |
| |
| Keywords: | speech recognition wavelet neural network genetic algorithm GA-WNN |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
