| 基于脉冲耦合神经网络的孤立词语音识别研究 |
| |
| 引用本文： | 刘琨,金文标.基于脉冲耦合神经网络的孤立词语音识别研究[J].重庆邮电大学学报(自然科学版),2008,20(2):217-220. |
| |
| 作者姓名： | 刘琨 金文标 |
| |
| 作者单位： | 重庆邮电大学,计算机学院,重庆,400065;河南理工大学,计算机学院,河南,焦作,454000;重庆邮电大学,计算机学院,重庆,400065 |
| |
| 摘 要： | 根据语音信号的“可视”特点，提出了一种基于简化脉冲耦合神经网络（PCNN）实现孤立词语音识别的方法。将语谱图输入到PCNN中得到相应的时间序列标识，作为语音信号的特征参数，然后通过模板匹配法和概率神经网络（PNN）相结合的方法实现语音分类识别。仿真结果表明，该方法能够达到较高的语音识别率。
|
| 关 键 词： | PCNN 语谱图 孤立词 语音识别 |
| 文章编号： | 1673-825X(2008)02-0217-04 |
| 收稿时间： | 2007/12/6 0:00:00 |
| 修稿时间： | 2007年12月6日 |
Research of isolated digit speech recognition based on pulse coupled neural network |
| |
| LIU Kun,JIN Wen-biao.Research of isolated digit speech recognition based on pulse coupled neural network[J].Journal of Chongqing University of Posts and Telecommunications,2008,20(2):217-220. |
| |
| Authors: | LIU Kun JIN Wen-biao |
| |
| Institution: | College of Computer Science and Technology, Chongqing University of Posts and Telecommunications, Chongqing 400065, P.R.China |
| |
| Abstract: | A novel approach for isolated digit speech recognition based on simplified pulse coupled neural network (PCNN) is presented in terms of the visible characteristics of speech. PCNN is employed to the spectrogram to produce the corresponding time series icon as the feature parameters of speech. Then the isolated speech is classified and recognized by means of template matching and probabilistic neural network. Simulation results validate the efficiency of the proposed algorithm. |
| |
| Keywords: | pulse coupled neural network (PCNN) spectrogram isolated digit speech recognition |
| 本文献已被 维普 万方数据 等数据库收录！ |
| 点击此处可从《重庆邮电大学学报(自然科学版)》浏览原始摘要信息 |
| 点击此处可从《重庆邮电大学学报(自然科学版)》下载免费的PDF全文 |
