摘要
对声音识别技术进行了深入研究,提出一种声音识别传感器设计。该传感器首先采集声音信号,经过整形、放大后进行A/D转换,提取语音特征,并利用动态时间规整(DTW)算法进行识别;传感器采用C/S架构和外部设备进行通信,通过串口接收外部设备命令,分析处理后传回识别结果。最后设计并实现智能家居硬件环境,通过声音识别传感器完成智能家居的远程遥控,完成诸如开灯、关灯等动作。实验结果表明:该传感器工作稳定,识别率高,能够应用于各种场合。
Voice recognition technology is deeply studied and a design of voice recognition sensor is proposed. The sensor aquires voice signal firstly, through shaping, amplification and A/D conversion, extract phonetic feature and use DTW algorithm to identify. The sensor uses C/S architecture to communicate with external devices, receives external device command via serial port, analyzes, processes and returns recognition results. Finally, hardware design for intelligent home environment is accomplished to complete remote control of intelligent home such as turning on/off the lights and other activities, through voice recognition sensor. Experimental results show that the sensor is stable,and has high recognition rate,which can be applied to a variety of occasions.
出处
《传感器与微系统》
CSCD
北大核心
2014年第12期51-53,共3页
Transducer and Microsystem Technologies
基金
国家"863"高技术研究发展计划资助项目(2009AA011700)
国家自然科学基金青年基金资助项目(61302158)
江苏省自然科学基金资助项目(BK20130869)
关键词
嵌入式系统
智能家居
声音识别传感器
embedded system
intelligent home
voice recognition sensor
