嵌入式语音识别呼救装置的设计与实现  

Design and realization of embedded speech recognition annunciator

作　　者：沈音乐[1] 

机构地区：[1]杭州职业技术学院,浙江杭州310018

出　　处：《计算机工程与设计》2008年第10期2526-2529,共4页Computer Engineering and Design

基　　金：浙江省教育厅2006年科研基金项目(20061728)

摘　　要：阐述了基于语音识别和多媒体无线远程传输两项技术的实时呼救系统的设计思路,介绍了其中的几个关键性技术要点。系统以ARM为硬件平台,使用嵌入式Linux操作系统进行开发,完成了用呼救音触发呼救监控装置,及时将多媒体呼救信息传给救援方,为居家安防提供了一种解决方案和技术依据。A method based on speaker-dependent speech recognition and multimedia message transmition is described and some Key technologies are introduced. The system, which is based on arm CPU and using embedded-Linux as operating system, can realize sending multimedia message to ask for help with the annunciator provoked by voice. A solution example is provided as reference and technology support for the development of house safeguard.

关 键 词：语音识别 隐马尔科夫模型 梅尔频率倒谱系数 多媒体信息服务 呼救装置 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]