用于管制语音理解的语义分析方法

详细信息    查看全文 | 推荐本文 |

英文篇名：Semantic Analysis Method for Control Voice Understanding 作者：王煊 ; 王冠 ; 蒋伟煜 ; 徐秋程 ; 付胜豪 英文作者：WANG Xuan;WANG Guan;JIANG Weiyu;XU Qiucheng;FU Shenghao;State Key Laboratory of Air Traffic Management System and Technology;The 28th Research Institute of China Electronics Technology Group Corporation; 关键词：结构化模板 ; 管制指令语义网 ; 自动语音识别 ; 自然语言处理 英文关键词：control instruction;;structural template;;semantic network;;automatic speech recognition;;natural language processing 中文刊名：ZHXT 英文刊名：Command Information System and Technology 机构：空中交通管理系统与技术国家重点实验室;中国电子科技集团公司第二十八研究所; 出版日期：2019-03-20 15:55 出版单位：指挥信息系统与技术 年：2019 期：v.10;No.55 基金：国家重点研发计划(2016YFB0502405)资助项目 语种：中文; 页：ZHXT201901006 页数：5 CN：01 ISSN：32-1818/TP 分类号：36-40

摘要

管制指令语义理解对场面冲突判断至关重要,为了避免因管制员失误导致的场面冲突问题,需采用自动语音识别和自然语言处理技术在第一时间提取管制指令并进行理解。以无线电陆空通话规则及其基本语法理论为基础,设计了一种管制指令结构化模板。为了更细致地理解管制指令中动词与其他词之间的关系,基于格语法理论构建了管制指令语义网,并建立了动词-论元-关系三元组。该模板可用于解析实际管制指令,从而实现计算机自动理解管制指令功能。为了将管制语音转化为结构化管制指令,采用了自动语音识别和自然语言处理技术。试验结果表明,该模板可用于计算机对管制指令的自动理解。
        The semantic understanding of control instruction is crucial to the scene conflict judgment. To avoid the scene conflicts caused by the controller′s fault, the automatic speech recognition(ASR) and the natural language processing(NLP) can be used to extract the control instruction and understand them in the first time. Based on the ground-air radio communication rules ant its basic grammar theory, a control instruction structural template is designed. To understand in greater detail the relationship between verbs and other words in a control instruction, the control instruction semantic network is constructed based on the case grammar theory. The verb-argument-relation triad is established. The template can be used to parse actual control instruction and realize the function of automatic understanding of the control instruction. The automatic speech recognition and natural language processing are used to convert the control voice to the structural control instruction. The experimental results show that the template can be used to the automatic understanding of control instruction.

引文

[1]严勇杰,曹罡.下一代空管系统运行概念及其关键技术[J].指挥信息系统与技术,2018,9(3):8-17.
    [2]何昕,蒋豪,韩丹.管制指令特征参数提取研究[J].科学技术与工程,2015,15(20):89-94.
    [3]何昕,韩丹.管制指令语音识别模板库的关键技术研究[J].中国西部科技,2015,14(3):1-3.
    [4]李春利,董冠灵,郭倩.基于HMM的空管指令语音识别研究[C]//智能运输系统学术研讨会.天津:中国民航大学,2007.
    [5]袁伟伟,胡军,刘万凤.空管模拟训练中指令的语音识别与合成技术研究[J].电子设计工程,2013,21(6):8-11.
    [6]王蓓蓓,王匀.基于管制语音识别的场监系统离港航空器自动标识[J].电子质量,2014(7):46-49.
    [7]陈晨,向东.基于管制语音特点统计分析的管制负荷评价[J].空中交通管制,2010(10):4-6.
    [8]杨金锋,卢薇冰,贾桂敏,等.基于深度CNN的陆空通话语义一致性校验[J].中国民航大学学报,2018,36(1):37-41.
    [9]KOPALD H D,CHANEN A,CHEN S,et al.Applying automatic speech recognition technology to air traffic management[C]//IEEE/AIAA 32nd Digital Avionics Systems Conference(DASC).New York:IEEE,2013:1-11.
    [10]FERNáNDEZ F,FERREIROS J,PARDO J M,et al.Automatic understanding of ATC speech[J].IEEEAerospace and Electronics Systems Magazine,2006,21(10):12-17.
    [11]SCHMIDT A,OUALIL Y,OHNEISER O,et al.Context-based recognition network adaptation for improving on-line ASR in air traffic control[C]//2014IEEE Spoken Language Technology Workshop(SLT).[S.l]:IEEE,2014:13-18.
    [12]CRAPARO E,FERON E.Natural language processing in the control of unmanned aerial vehicles[C]//AIAAGuidance,Navigation,and Control Conference and Exhibit.Providence:AIAA,2004:1-13.
    [13]王万乐,赵琦.空中交通管制员无线电陆空通话[M].北京:北京交通大学出版社,2016.
    [14]刘颖.计算语言学[M].修订版.北京:清华大学出版社,2014.
    [15]袁毓林.汉语配价语法研究[M].北京:商务印书馆,2010.
    [16]菲尔墨C J.“格”辨[M].胡明扬,译.北京:商务印书馆,2012.
    [17]中国民用航空总局.空中交通无线电通话用语指南[M].成都:西南交通大学出版社,2005.
    [18]GENNARI J H,MUSEN M A,FERGERSON W,et al.The evolution of Protégé:an environment for knowledge based systems development[J].International Journal of Human-Computer Studies,2009,58(1):89-123.