基于铁路调度指挥作业指令的关键词识别技术应用研究  

Application research on keyword recognition technology based on railway dispatching operation instruction

作　　者：杨博 何振华 冯立恒 郭星 毛疆华 YANG Bo;HE Zhenhua;FENG Liheng;GUO Xing;MAO Jianghua(Dispatching office of China Railway Beijing Bureau Group Co.Ltd.,Beijing100036,China;Zhengzhou Xinda Advanced Technology Research Institute,Zhengzhou450000,China;Henan Xinda Yukun Intelligent Technology Co.Ltd.,Zhengzhou450000,China)

机构地区：[1]中国铁路北京局集团有限公司调度所,北京100036 [2]郑州信大先进技术研究院,郑州450000 [3]河南信大煜坤智能科技有限公司,郑州450000

出　　处：《铁路计算机应用》2022年第8期34-39,共6页Railway Computer Application

基　　金：中国铁路北京局集团有限公司重大课题(2021AY02)。

摘　　要：针对规范铁路调度人员的口头指令这一问题,计划使用关键词识别技术来对调度人员的作业指令进行识别记录,对不规范用语进行约束调整。采用基于关键词–废料模型的关键词识别技术,使用非词格依赖的最大互信息准则端到端训练方法,进行关键词识别模型训练,实现对铁路调度人员口头指令的自动识别。针对出现频率最高的2个关键词进行实验分析,研究该方法在工业场景下应用的可行性。实验结果表明,通过调节参数控制误识率,可达到一定的关键词识别率,具备应用可行性。In order to standardize the verbal instructions of railway dispatchers,this paper planned to use keyword recognition technology to identify and record the operation instructions of dispatchers,and restrict and adjust the nonstandard terms.Based on the recognition technology of keyword-waste model,the paper used the end-to-end training method of maximum mutual information criterion without word lattice dependence to train the keyword recognition model,so as to realize the automatic recognition of railway dispatcher's oral instructions.Experimental analysis was carried out for the two keywords with the highest frequency to study the feasibility of this method in industrial scenarios.The experimental results show that by adjusting parameters to control the error recognition rate,a certain keyword recognition rate can be achieved which is feasible for application.

关 键 词：铁路调度作业指令 语音识别 关键词识别 唤醒词检测 非词格依赖最大化互信息准则 

分 类 号：U292.4[交通运输工程—交通运输规划与管理]