一种特定人语音家居控制系统.pdfVIP

Computer Science and Application 计算机科学与应用, 2017, 7(10), 935-943 Published Online October 2017 in Hans. /journal/csa /10.12677/csa.2017.710106 A Specific Speech Control Home System Yan Peng, Jingkuo Zhang, Runxue Chen, Shuang Ren School of Computer, Sichuan University of Science Engineering, Zigong Sichuan th th th Received: Sep. 28 , 2017; accepted: Oct. 10 , 2017; published: Oct. 17 , 2017 Abstract The application of speech control in smart home brings great convenience to life. However, con- ventional independent speech control will make incorrect speech triggering. Based on the non- specific speech recognition, this paper designs a specific human voice recognition home control system by improving the corresponding algorithm. This system gets speech features by MFCC al- gorithm and matches speech feature by DTW algorithm. Host computer is based on Android plat- form. This paper uses Android Studio tools and NDK assembly to achieve the development of software and algorithm. Lower computer is based on Zigbee platform used IAR to achieve lower computer development of smart home. As a result, this system can solve specific speech control with a higher recognition rate and more human operation. Keywords Home System, Specific Speech, Speech Control, Android Platform, Zigbee Platform 一种特定人语音家居控制系统 彭 龑，张景扩，陈闰雪，任 爽 四川理工学院计算机学院，四川 自贡 收稿日期：2017年9月28 日；录用日期：2017年10月10 日；发布日期：2017年10月17 日 摘 要 语音控制在智能家居的应用为人民生活带来极大便利，但常规的非特定人语音控制会产生语音的误触发 [1]。本文在非特定语音识别的基础上，通过改进相应算法设计了一种特定人语音识别家居控制系统。系 统采用MFCC算法获取语音特征，通过DTW算法实现语音特征匹配。上位机基于安卓平台，利用Android Studio工具并结合NDK组件实现相应软件与算法开发。下位机基于Zigbee平台，利用IAR完成家居系统 下位机的开发。系统可完成特定人的语音控制，具有较高的识别率，操作更加人性化。 文章引