电力物资质量检测环节语音信号预加重处理方法  

Pre-emphasis processing method of voice signal in power material quality inspection link of electric power materials

在线阅读下载全文

作  者:张艳 李杨 林明晖 董元龙 钱锡颖 ZHANG Yan;LI Yang;LIN Minghui;DONG Yuanlong;QIAN Xiying(Ningbo New Victory Electric Co.,Ltd.,Ningbo 315000,Zhejiang China;State Grid Ningbo Electric Power Supply Company,Ningbo 315000,Zhejiang China;Ningbo Electric Power Transmission and Substation Construction Co.,Ltd.,Ningbo 315032,Zhejiang China)

机构地区:[1]宁波新胜中压电器有限公司,浙江宁波315000 [2]国网浙江省电力有限公司宁波供电公司,浙江宁波315000 [3]宁波送变电建设有限公司,浙江宁波315032

出  处:《粘接》2024年第11期183-185,189,共4页Adhesion

摘  要:针对电力物资质量检测环节中的语音交互式检测信息容易受到周围环境因素的影响问题,提出声音信号预加重方法研究。构建声音信号数学模型,分析其时域特征,以此为基础,计算每帧声音信号频谱能量与概率分布,衡量每帧声音信号谱熵值,检测声音信号端点,估计声音信号先验信噪比推出增益函数,以端点检测结果为界限,以增益函数为依据,采用重叠相加运算处理声音信号,实现声音信号的预加重。实验数据显示,应用提出方法预加重处理声音信号后,其平均信噪比数值最大值为20.15 dB,充分证实了提出方法信号预加重效果较佳。In order to solve the problem that the voice interactive detection information in the quality inspection process of power materials is easily affected by the surrounding environmental factors,a pre-emphasis method of sound signal was proposed.The mathematical model of the sound signal was constructed,and its time domain characteristics were analyzed,and on this basis,the spectral energy and probability distribution of each frame of the sound signal were calculated,the spectral entropy value of each frame of the sound signal was measured,the endpoint of the sound signal was detected,and the prior signal-to-noise ratio of the sound signal was estimated to introduce the gain function.The experimental data showed that the maximum value of the average signal-to-noise ratio was 20.15 dB after applying the proposed method to pre emphasis the sound signal,which fully proves that the proposed method has better signal pre emphasis effect.

关 键 词:质量检测环节 检测信息 语音交互 电力物资 预加重 

分 类 号:TP274.4[自动化与计算机技术—检测技术与自动化装置] TM711.2[自动化与计算机技术—控制科学与工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象