语音技术在塔台模拟机上的应用  

Application of voice technology in tower simulator

作　　者：胡誉 Hu Yu(Air Traffic Management Bureau CAAC,Guangzhou 510000,China)

机构地区：[1]中国民用航空中南地区空中交通管理局,广东广州510000

出　　处：《无线互联科技》2020年第4期5-6,共2页Wireless Internet Technology

摘　　要：基于空管模拟训练机对空管人员训练的重要性和优越性以及塔席特点,利用计算机实现飞行器动态特征的场景再现,设定多场景、多任务的训练科目是目前采用的普遍做法。文章依托塔台模拟机性能特点,采用有限词汇量识别方法,结合空管发音规则、语法结构、发音标准以及中英双语等特点,采用动态时间归正识别技术和语音同步叠加算法实现语音编码、识别与合成,实现了多局部最优化决策。Based on the importance and superiority of air traffic control simulation training machine for air traffic control personnel training and the characteristics of tower seat,it is a common practice to use computer to realize the scene reproduction of aircraft dynamic characteristics and set multi scene and multi task training subjects in this paper.Based on the performance characteristics of the tower simulator,the limited vocabulary recognition method is adopted,combined with the characteristics of air traffic control pronunciation rules,grammatical structure,pronunciation standards and Chinese English bilingualism,and the dynamic time correction recognition technology and speech synchronous superposition algorithm are adopted to realize speech coding,recognition and synthesis,and multi local optimal decision-making is realized.

关 键 词：塔台模拟机 语音技术 动态时间规整识别 PSOLA算法 内容扰乱系统 

分 类 号：TN912.3[电子电信—通信与信息系统] V355.1[电子电信—信息与通信工程]