基于遗传算法和小波神经网络的语音识别研究  

Research of speech recognition based on genetic algorithm and wavelet neural network

作　　者：斯芸芸[1] 徐道连[1] 周卓然[1] 

机构地区：[1]重庆大学光电工程学院,重庆400030

出　　处：《微型机与应用》2011年第16期68-71,共4页Microcomputer & Its Applications

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。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 performance and global optimization. Therefore, we can combine genetic algorithm and wavelet neural network to form a hybrid algorithm to train the neural network GA-WNN algorithm. The simulation results show that the new algorithm can effectively shorten the recognition time, increase the speed of network training recognition, and improve speech recognition rate.

关 键 词：语音识别 小波神经网络 遗传算法 GA—WNN 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]