基于声音采集与分类技术的报警系统设计

doi:10.16257/j.cnki.1681-1070.2022.1115

电子与封装 ›› 2022, Vol. 22 ›› Issue (11): 110502 . doi: 10.16257/j.cnki.1681-1070.2022.1115

基于声音采集与分类技术的报警系统设计

徐婕;沈丹丹;潘硕;朱习松

中科芯集成电路有限公司，江苏无锡 214072

收稿日期:2022-05-18 出版日期:2022-11-29 发布日期:2022-08-17
作者简介:徐婕（1988—），女，陕西延安人，硕士，工程师，主要研究方向为系统设计与质量管理。

Alarm System Design Based on Sound Acquisition and Classification Technology

XU Jie, SHEN Dandan, PAN Shuo, ZHU Xisong

China Key System& Integrated Circuit Co., Ltd., Wuxi 214072, China

Received:2022-05-18 Online:2022-11-29 Published:2022-08-17

摘要/Abstract

摘要： 环境声音包含丰富的信息，通过声音识别可以快速判断异常信息，从而为异常确认、快速处理争取时间。给出了一种基于FPGA+ADC的声音报警系统，语音芯片XX73311完成模拟采集，以太网PHY电路XX83848实现数据传输。系统可高质量完成模拟量到数字量的转换，并以短时能量端点检测、特征参数提取为基础训练样本，以声音识别技术完成异常声音判断，实现异常报警。试验结果表明，使用声音识别具有良好的实际效果，报警系统能够完成目标任务。

关键词: 环境声音, 声音识别, 报警系统, 异常报警

Abstract: Environmental sound contains a wealth of information. Through sound recognition, abnormal information can be quickly judged, so as to buy time for abnormal confirmation and rapid processing. A sound alarm system based on FPGA+ADC is presented. The analog acquisition is completed by voice chip XX73311, and the data transmission is realized by Ethernet PHY circuit XX83848. The system can complete the conversion from analog quantity to digital quantity with high quality. It trains samples based on short-term energy endpoint detection and feature parameter extraction, and uses sound recognition technology to judge the abnormal sound to realize abnormal alarm. The test results show that the use of sound recognition has good practical results, and the alarm system can complete the target task.

Key words: environmental sound, sound recognition, alarm system, abnormalalarm

中图分类号:

TN912.3

徐婕;沈丹丹;潘硕;朱习松. 基于声音采集与分类技术的报警系统设计[J]. 电子与封装, 2022, 22(11): 110502 .

XU Jie, SHEN Dandan, PAN Shuo, ZHU Xisong. Alarm System Design Based on Sound Acquisition and Classification Technology[J]. Electronics & Packaging, 2022, 22(11): 110502 .

参考文献

[1] 李伟，李硕. 理解数字声音——基于一般音频/环境声的计算机听觉综述[J]. 复旦学报(自然科学版), 2019, 58(3): 269-313.
[2] ZIEGER C, BRUTTI A, SVAIZER P. Acoustic based surveillance system for intrusion detection[C]// Sixth IEEE International Conference on Advanced Video & Signal Based Surveillance. IEEE, 2009.
[3] 孙继平，余星辰. 基于声音识别的煤矿重特大事故报警方法研究[J]. 工矿自动化, 2021, 47(2): 1-5, 44.
[4] 谢杰，陈政，傅建军. 基于FPGA的车牌定位系统研究[J]. 电子与封装, 2022, 22(1): 010502.
[5] 宋知用. MATLAB在语音信号分析与合成中的应用[M]. 北京：北京航空航天大学出版社, 2013.
[6] COWLING M, SITTE R. Comparison of techniques for environmental sound recognition[J]. Pattern recognition letters, 2003, 24(15): 2895-2907.
[7] VERGARA L, MORAGUES J, GOSáLBEZ J, et al. Detection of signals of unknown duration by multiple energy detectors[J]. Signal Processing, 2010, 90(2): 719-726.
[8] 陈迪，龚卫国，杨利平. 基于基音周期的语音MFCC参数提取[J]. 计算机应用, 2007, 27(5): 1217-1219.
[9] RABINER L R, JUANG B H. Fundamentals of speech recognition[M]. Upper Saddle River: Prentice Hall, 1993: 4.
[10] 刘敬伟，徐美芝，郑忠国，等. 基于DTW的语音识别和说话人识别的特征选择[J]. 模式识别与人工智能, 2005, 18(1): 50-54.
[11] GERHARD D. Audio signal classification: History and current techniques[M]. Regina: Department of Computer Science, University of Regina, 2003.
[12] CHANDRAKALA S, SEKHAR C C. Model based clustering of audio clips using Gaussian mixture models[C]// Seventh International Conference on Advances in Pattern Recognition, 2009: 47-50.
[13] FREEMAN G, DONY R D, AREIBI S M. Audio environment classication for hearing aids using artificial neural networks with windowed input[C]// Computational Intelligence in Image and Signal Processing, 2007: 183-188.
[14] PAN W J, YAO Y, LIU Z J, et al. Audio classification in a weighted SVM[C]// International Symposium on Communications and Information Technologies, 2007: 468-472.
[15] 刘波霞，陈建峰. 基于特征分析的环境声音事件识别算法[J]. 计算机工程, 2011，37(22): 261-263.

中国半导体行业协会封装分会会刊

中国电子学会电子制造与封装技术分会会刊

基于声音采集与分类技术的报警系统设计

Alarm System Design Based on Sound Acquisition and Classification Technology

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 2

编辑推荐

Metrics

本文评价

[1]	苗韵, 鲍宜鹏, 杨晓刚. 一种HDMI-CEC控制器的设计与实现[J]. 电子与封装, 2020, 20(2): 20304-.
[2]	刘云晶，刘梦影. 一种32位MCU的FPGA验证平台[J]. 电子与封装, 2020, 20(1): 10205-.