基于声纹识别技术的Android设备声音功能自动化测试系统  

An automatic testing system for audio function of Android devices based on the voiceprint recognition technology

作　　者：李嘉伟[1] 胡海龙[1] 林志贤[1] Li Jiawei;Hu Hailong;Lin Zhixian(College of Physics and Information Engineering,Fuzhou University,Fuzhou 350116,China)

机构地区：[1]福州大学物理与信息工程学院,福建福州350116

出　　处：《信息技术与网络安全》2018年第5期106-109,共4页Information Technology and Network Security

基　　金：国家重点研发计划(2016YFB0401503);福建省科技重大专项(2014HZ0003-1);广东省科技重大专项(2016B090906001);福建省资助省属高校专项课题(JK2014002)

摘　　要：显示器生产测试过程中需要对设备扬声器与麦克风等声音功能品质进行测试,针对现有测试系统自动化程度较低的问题,提出一种基于声纹识别技术的Android设备声音功能自动化测试系统。本系统采用MFCC+ΔMFCC算法提取语音信号特征,通过DTW算法对麦克风录制语音和扬声器播放语音的模型特征进行比对,实现对扬声器与麦克风同步的失真、无声、杂声等多种不良情况的有效检测。本系统良品与无声不良检测准确率接近100%,失真不良检测准确率高达96.67%,杂声不良检测准确率达93.33%。该方案相比人工测试效率提高6.25倍,且成本低、容易实现,有助于提升企业生产效率。During the production testing of the display devices,it is necessary to test the audio function quality of the speaker and microphone.Aiming at the low automation and applicability of the existing test system,we propose an automatic testing system for the audio function of Android devices based on the voiceprint recognition technology. This system uses MFCC ＋ ΔMFCC algorithm to extract the characteristic value of speech signal,and compares the model characteristic value of voice recorded by microphone and broadcasted by loudspeaker using DTW algorithm,so as to effectively detect the various defects of the speaker and microphone simultaneously,such as distortion,soundlessness,and noise. The detection accuracy of the system is nearly 100% for non-defective and soundless products,96. 67% for speech distortion,and 93.33% for noise detection. Compared with the manual testing,the efficiency can be increased by 6. 25 times. This system is of low cost and facility,and conductive to improve the production efficiency of the enterprise.

关 键 词：ANDROID 声纹识别 MFCC DTW 自动化测试系统 

分 类 号：TP277[自动化与计算机技术—检测技术与自动化装置] TN912.3[自动化与计算机技术—控制科学与工程]