基于语音识别的智能家居系统研究  

Research of smart home system based on speech recognition

作　　者：刘其洪[1] 李仲阳[1] 徐孟龙[1] 

机构地区：[1]湖南师范大学工学院,湖南长沙410081

出　　处：《信息化纵横》2009年第14期80-83,共4页

基　　金：湖南省教育厅科研资肋项目(08B050)

摘　　要：针对智能家居系统中信息不能随时随地进行控制及其交流方式不流畅的问题,融合电话公用网和家庭网络设计了一个基于ARM9和语音识别技术的智能家居系统。该系统采用RASTA滤波方法去除语音信号中夹杂的卷积信道噪声,采用改进的动态时间规正(DTW)算法对语音命令进行识别。A smart home system based on ARM9 and speech recognition technology was designed,which merged public telephone network and home network, and made smart home system be controlled at any time and any where and the communicating way be natural and smooth. In the system, RelAtive SepeTrAI（RASTA） filter was used to reduce the convolution channel noise mixed in speech signal, and the improved dynamic time warping （DTW） algorithm was adopted for recognizing speech command.

关 键 词：智能家居 语音识别 带通滤波器 动态时间规正 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]