增强型声源定位,英文

1526IEEE TRANSACTIONS ON SYSTEMS,MAN,AND CYBERNETICS—PART B:CYBERNETICS,VOL.34,NO.3,JUNE2004

Enhanced Sound Localization

Bob Mungamuru and Parham Aarabi,Member,IEEE

Abstract—A new approach to sound localization,known as en-hanced sound localization,is introduced,offering two major ben-efits over state-of-the-art algorithms.First,higher localization ac-curacy can be achieved compared to existing methods.Second,an estimate of the source orientation is obtained jointly,as a conse-quence of the proposed sound localization technique.The orienta-tion estimates and improved localizations are a result of explicitly modeling the various factors that affect a microphone’s level of ac-cess to different spatial positions and orientations in an acoustic environment.Three primary factors are accounted for,namely the source directivity,microphone directivity,and source-microphone distances.

Using this model of the acoustic environment,several different enhanced sound localization algorithms are derived.Experiments are carried out in a real environment whose reverberation time is 0.1seconds,with the average microphone SNR ranging between 10–https://www.sodocs.net/doc/3a13889599.html,ing a24-element microphone array,a weighted ver-sion of the SRP-PHAT algorithm is found to give an average local-ization error of13.7cm with3.7%anomalies,compared to14.7cm and7.8%anomalies with the standard SRP-PHAT technique. Index Terms—Microphone arrays,sound localization,source tracking.

I.I NTRODUCTION

F ACILITATED by the staggering increase in the availability

of computational power over the last two decades,a new focus in the research community has emerged,known as perva-sive computing.The field of pervasive computing encompasses the rapidly expanding class of computers distributed and em-bedded within larger devices classically not viewed as being mi-croprocessor driven.

One of the key technological issues in the development of pervasive systems is designing improved,advanced methods of human–computer interaction(HCI)[7],[8],[15].Speech and physical gestures as interface modalities seem to be the heirs apparent to the tethered keyboard and mouse that the world has grown accustomed to.Human–computer interactions in this new pervasive era promise a much more mobile,natural,interactive and spatially rich computing experience heretofore unseen with standard desktop machines.

In HCI systems offering speech interfaces,determination of the physical position and orientation of users is an essential ca-pability[5],[6],[9].Without precise knowledge of the spatial location of users in an environment,the interface would not be able to react naturally to the needs and behaviors of the user. Many other applications exist that would be aided by ac-curate user localization and orientation estimation.Examples

Manuscript received August22,2003;revised December27,2003.This paper was recommended by Associate Editor C.-T.Lin.

The authors are with the Department of Electrical and Computer En-gineering,University of Toronto,Toronto,ON,M5S364Canada(e-mail: parham@ecf.utoronto.ca).

Digital Object Identifier10.1109/TSMCB.2004.826398include intelligent living environments,speech separation for hands-free communication devices such as personal digital as-sistants,and security systems[7],[8],[15].These would all ben-efit from the spatial information and speech enhancement made available by robust user localization and orientation estimation techniques.

A common method of localizing a user is by means of sound source localization[1],[4],[5],[6],[9],[10],[12].Sound lo-calization is the process of determining the spatial location of a sound source based on multiple observations of the emitted sound signal.

Current sound localization techniques are generally based upon the idea of computing the delay-and-sum energy of an array of observations,and selecting the position in space that yields a maximum.This can be done with just two micro-phones,or be extended to larger arrays.Closely related to this idea is filter-and-sum beamforming[9],whereby each microphone’s observation is filtered in some advantageous way,and the results are then summed.Once again,the spatial position yielding maximum energy is selected.The choice of filtering strategy in filter-and-sum beamforming is driven by the specifics of the application.

A drawback of the existing techniques is that,when applied to large microphone arrays,the inherent reliability of the observed information is not taken into account.That is,the fact that cer-tain microphones are better suited than others to gather informa-tion on a given spatial position,is not reflected in the algorithm. For example,suppose a given microphone is much further away from a position of interest than another microphone.Clearly, the information provided by the nearby microphone should be given more importance than the one further away.Most existing techniques,however,do not do this.Instead,all microphones are considered equal,and the sound localization accuracy is reduced as a result.There exists a technique[2]that does weight different microphones according to their level of access to different spa-tial positions.The problem,however,is that the weighting is dependent on spatial position only,whereas,in reality other fac-tors do impact heavily.Proper modeling of the various factors is required to remedy this.

In this paper,a maximum likelihood microphone weighting algorithm for sound source localization is proposed.The deriva-tion of these weights requires knowledge about the orientation of the source,which is a parameter that is jointly estimated along with the source location.

II.S OUND L OCALIZATION

Several microphone array-based sound localization tech-niques have been proposed in the literature[1],[4],[5],[6],[9], [10],[12],with varying degrees of accuracy and computational complexity.Of these techniques,the most successful and

1083-4419/04$20.00?2004IEEE

MUNGAMURU AND AARABI:ENHANCED SOUND LOCALIZATION 1527

prevalent ones have generally been based upon the idea of

maximizing the steered response power (SRP),a quantity that will be described shortly.Different variations on this concept have led to algorithms with improved performance,to varying extents.Other approaches to sound localization include mul-tiple signal classification (MUSIC)[11],[17],and maximum likelihood (ML)estimation [18],although these are typically applied to far-field narrow band direction-of-arrival estimation problems.

A.Sound Localization Fundamentals

Suppose the source is emitting a sound signal,,and a two-microphone array is observing signals and .The signal received by the microphones will be distorted due to both the acoustics of the room and the ambient noise in the environment.Furthermore,due to the distance between the two microphones in the array,there will be a measurable time dif-ference between observations of the signal at each microphone.This is referred to as the time difference of arrival (TDOA),,between the microphones.If the value of can be deter-mined somehow,the location of the sound source would be con-strained to the set of all points in space corresponding to the given TDOA.This set of points traces out a hyperbola in two-di-mensional (2-D)space.

A TDOA estimate can be obtained by finding that maxi-mizes the cross correlation between the two microphone signals.The cross correlation is normally computed in the frequency do-main,so that noninteger values of can be tested:

(1)

where

and are the Fourier transforms of and ,and a multiplying constant of has been dropped.

An alternative way of viewing this maximization would be

to assign a likelihood

to each position in the room,and then decide that the highest likelihood corresponds to the correct position for the source

(2)

where the position is the one that maximizes

.We would therefore,steer the microphone array to each position in the room,by finding the TDOA corresponding to the given position and assigning a likelihood proportional to the computed cross correlation:

(3)

where

is the TDOA corresponding to position .Since the delay-and-sum energy is being computed at each position in the room,this quantity is referred to as the SRP [9].

The function

is a map of likelihoods for each position in space,called a spatial likelihood function (SLF)[2].The SLF corresponding to the two-microphone case discussed here,is shown in Fig.1.

Fig.1.

Two-microphone SLF.

The simple two-microphone SRP-based localization scheme can be extended easily to the -microphone case.For each point in space,the corresponding vector of time delays,,is found and the cross correlation is computed for each pairwise com-bination of microphones.These pairwise correlations are then simply added up:

(4)

where and are the th and th elements of the vector of time

delays

.It can be shown that adding up the pairwise SRPs is still a delay-and-sum beamforming technique.B.Generalized Cross Correlation Techniques

When estimating TDOAs using the integral in (1),the relative importance of each frequency in the cross correlation integral is given by the magnitude of cross power spectrum at that fre-quency.This is irrespective of the actual characteristics of the signals.

If,however,additional application-specific information re-garding the signals was available,it might be better to weight the frequencies differently.To this end,the generalized cross correlation (GCC)[13]may be substituted in place of the stan-dard cross correlation in (4):

(5)

where is a frequency-dependent weighting function that determines the relative importance of each frequency in esti-mating the time-delay vector

.Whereas the standard cross correlation in (4)tries to max-imize the delay-and-sum energy,the GCC tries to maximize the filter-and-sum energy.In this respect,the weight function

can be viewed as a prefilter that is applied to the ob-served pairwise cross-spectra,before carrying out the standard delay-and-sum energy maximization.

The choice of the function

has been studied extensively for sound sources,and three common choices will be described

常见职务、职位英文翻译

常见职位、职务英文译名 Accounting Assistant 会计助理 Accounting Clerk 记帐员 Accounting Manager 会计部经理 Accounting Stall 会计部职员 Accounting Supervisor 会计主管 Administration Manager 行政经理 Administration Staff 行政人员 Administrative Assistant 行政助理 Administrative Clerk 行政办事员 Advertising Staff 广告工作人员 Airlines Sales Representative 航空公司定座员 Airlines Staff 航空公司职员 Application Engineer 应用工程师 Assistant Manager 副经理 Bond Analyst 证券分析员 Bond Trader 证券交易员 Business Controller 业务主任 Business Manager 业务经理 Buyer 采购员 Cashier 出纳员 Chemical Engineer 化学工程师 Civil Engineer 土木工程师 Clerk/Receptionist 职员／接待员 Clerk Typist & Secretary 文书打字兼秘书 Computer Data Input Operator 计算机资料输入员 Computer Engineer 计算机工程师 Computer Processing Operator 计算机处理操作员 Computer System Manager 计算机系统部经理 Copywriter 广告文字撰稿人 Deputy General Manager 副总经理 Economic Research Assistant 经济研究助理 Electrical Engineer 电气工程师 Engineering Technician 工程技术员 English Instructor/Teacher 英语教师

基于MATLAB的声源定位系统

基于MATLAB的声源定位系统摘要 确定一个声源在空间中的位置是一项有广阔应用前景的有趣研究，将来可以广泛的应用于社会生产、生活的各个方面。 声源定位是通过测量物体发出的声音对物体定位，与使用声纳、雷达、无线通讯的定位方法不同，前者信源是普通的声音，是宽带信号，而后者信源是窄带信号。根据声音信号特点，人们提出了不同的声源定位算法，但由于信号质量、噪声和混响的存在，使得现有声源定位算法的定位精度较低。此外，已有的声源定位方法的运算量较大，难以实时处理。 关键词：传声器阵列；声源定位；Matlab

目录 第一章绪论 (1) 第二章声源定位系统的结构 (2) 第三章基于到达时间差的声源定位原理 (3) 第四章串口通信 (5) 第五章实验电路图设计 (8)

第六章总结 (16) 第七章参考文献 (17) 第一章绪论 1.1基于传声器阵列的定位方法简述 在无噪声、无混响的情况下，距离声源很近的高性能、高方向性的单传声器可以获得高质量的声源信号。但是，这要求声源和传声器之间的位置相对固定，如果声源位置改变，就必须人为地移动传声器。若声源在传声器的选择方向之外，则会引入大量的噪声，导致拾取信号的质量下降。而且，当传声器距离声源很远，或者存在一定程度的混响及干扰的情况下，也会使拾取信号的质量严重下降。为了解决单传声器系统的这些局限性，人们提出了用传声器阵列进行声音处理的方法。

传声器阵列是指由一定的几何结构排列而成的若干个传声器组成的阵列。相对于单个传声器而言具有更多优势，它能以电子瞄准的方式从所需要的声源方向提供高质量的声音信号，同时抑制其他的声音和环境噪声，具有很强的空间选择性，无须移动传声器就可对声源信号自动监测、定位和跟踪，如果算法设计精简得当，则系统可实现高速的实时跟踪定位。 传声器阵列的声音信号处理与传统的阵列信号处理主要有以下几种不同： （1）传统的阵列信号处理技术处理的信号一般为平稳或准平稳信号，相关函数可以通过时间相关来准确获得，而传声器阵列要处理的信号通常为短时平稳的声音信号，用时间平均来求得准确的相关函数比较困难。 （2）传统的阵列信号处理一般采用远场模型，而传声器阵列信号处理要根据不同的情况选择远场模型还是使用近场模型。近场模型和远场模型最主要的区别在于是否考虑传声器阵列各阵元因接收信号幅度衰减的不同所带来的影响，对于远场模型，信源到各阵元的距离差与整个传播距离相比非常小，可忽略不计，对于近场模型，信源到各阵元的距离差与整个传播距离相比较大，必须考虑各阵元接收信号的幅度差。 （3）在传统的阵列信号处理中，噪声一般为高斯噪声(包括白、色噪声)，与信源无关，在传声器阵列信号处理中噪声既有高斯噪声，也有非高斯噪声，这些噪声可能和信源无关，也可能相关。 由于上述阵列信号处理间的区别，给传声器阵列信号处理带来了极大的挑战。声波在传播过程中要发生幅度衰减，其幅度衰减因子与传播距离成正比，信源到传声器阵列各阵元的距离是不同的，因此声波波前到达各阵元时，幅度也是不同的。 另外，当声音信号在传播时，由于反射、衍射等原因，使到达传声器的声音信号的路径除了直达路径外还存在着多条其它路径，从而产生接收信号的幅度衰减、音质变差等不

常见职务职位英文翻译

常见职务职位英文翻译 希望对你有帮助哦!总公司Head Office分公司Branch Office营业部Business Office人事部Personnel Department(人力资源部)Human Resources Department总务部General Affairs Department财务部General Accounting Department销售部Sales Department促销部Sales Promotion Department国际部International Department出口部Export Department进口部Import Department公共关系Public Relations Department广告部Advertising Department企划部Planning Department产品开发部Product Development Department研发部Research and Development Department(R&D)秘书室Secretarial PoolAccounting Assistant 会计助理Accounting Clerk 记帐员Accounting Manager 会计部经理Accounting Stall 会计部职员Accounting Supervisor 会计主管Administration Manager 行政经理Administration Staff 行政人员Administrative Assistant 行政助理Administrative Clerk 行政办事员Advertising Staff 广告工作人员Airlines Sales Representative 航空公司定座员Airlines Staff 航空公司职员Application Engineer 应用工程师Assistant Manager 副经理Bond Analyst 证券分析员Bond Trader 证券交易员Business Controller 业务主任Business Manager 业务经理Buyer 采购员Cashier 出纳员Chemical Engineer 化学工程师

各类食品的英文翻译

麻油鸡汤：sesame chicken soup 排骨酥汤：spareribs soup 榨菜肉丝汤：Szechuan cabbage & pork soup 蛇羹(汤)：snake soup 甜点 爱玉：vegetarian gelatin 糖葫芦：tomatoes on stick 芝麻球：glutinous rice sesame ball 长寿桃：longevity peach 麻花：hemp flower 双胞胎：horse hooves 杏仁豆腐：almond tofu 豆花：uncongealed tofu 点心 小笼包：small steamed bun 豆沙包：smashed (red) bean bun 水饺：boiled dumpling 蒸饺：steamed dumpling 水晶饺：pyramid dumpling 年糕：new- years hard rice cake 元宵：full- moon dumpling (glutinous rice ball) 月饼：moon cake 茶叶蛋：boiled egg in tea 咸鸭蛋：salted duck egg 皮蛋：preserved egg (100-year egg) 凤梨酥：pineapple cake 锅贴：fried dumpling 蚵仔煎：oyster omelet 油豆腐：oily tofu (bean- curd) 臭豆腐：stinky tofu 甜不辣：tenpura 虾球：shrimp ball 春卷：spring roll 鸡卷：chicken roll 碗糕：salty rice pudding 筒仔米糕：rice tube pudding 红豆糕：red bean cake 萝卜糕：fried white radish cake 绿豆糕：bean paste cake 猪血糕：pork blood cake 芋头糕：taro cake 糯米糕：glutinous rice cake 肉圆：Taiwanese meatball (rice-meat dumpling) 当归鸭：angelica duck 加工食品 牛肉乾：dried beef 豆乾：dried tofu 冬菜：spiced cabbage 肉脯：fried pork crisps 肉松：crushed dried pork 面筋：flour gluten 豆瓣酱：broad bean paste 辣椒酱：chili sauce 泡菜：pickled vegetables (pickles) 榨菜：pickled mustard root 香肠：pork sausage 豆腐乳：preserved bean-curd (bean-curd cheese) 酱瓜：pickled cucumber 酱姜：soy-preserved ginger 萝卜乾：dried turnip 肉类及海鲜： 猪肉pork 猪排chop 五花肉streaky pork/marbled beef 肥肉fatty meat 瘦肉lean meat 前腿fore leg

常见职位职务英文翻译

常见职位职务英文翻译 Accounting Assistant会计助理 Accounting Clerk记帐员 Accounting Manager会计部经理 Accounting Stall会计部职员 Accounting Supervisor会计主管 Administration Manager行政经理 Administration Staff行政人员 Administrative Assistant行政助理 Administrative Clerk行政办事员 Advertising Staff广告工作人员 Airlines Sales Representative航空公司定座员 Airlines Staff航空公司职员 Application Engineer应用工程师 Assistant Manager副经理 Bond Analyst证券分析员 Bond Trader证券交易员 Business Controller业务主任 Business Manager业务经理 Buyer采购员 Cashier出纳员 Chemical Engineer化学工程师 Civil Engineer土木工程师 Clerk/Receptionist职员／接待员 Clerk Typist&Secretary文书打字兼秘书 Computer Data Input Operator计算机资料输入员Computer Engineer计算机工程师 Computer Processing Operator计算机处理操作员Computer System Manager计算机系统部经理 Copywriter广告文字撰稿人 Deputy General Manager副总经理 Economic Research Assistant经济研究助理 Electrical Engineer电气工程师 Engineering Technician工程技术员 English Instructor/Teacher英语教师 Export Sales Manager外销部经理 Export Sales Staff外销部职员 Financial Controller财务主任 Financial Reporter财务报告人 F.X.(Foreign Exchange)Clerk外汇部职员 F.X.Settlement Clerk外汇部核算员 Fund Manager财务经理 General Auditor审计长 General Manager/President总经理

一种改进的AEDA声源定位及跟踪算法

北京大学学报(自然科学版),第41卷,第5期,2005年9月 Acta Scientiarum Naturalium Universitatis Pekinensis,V ol.41,N o.5(Sept.2005) 一种改进的AEDA声源定位及跟踪算法1) 李承智　曲天书2)　吴玺宏 (北京大学视觉与听觉信息处理国家重点实验室,北京,100871;2)E2mail:qutianshu@https://www.sodocs.net/doc/3a13889599.html,) 摘　要　开展了基于麦克风阵列的真实声场环境声源定位的工作。针对传统的自适应特征值分 解时延估计算法收敛时间慢、对初值敏感以及不能有效跟踪时延变化等问题,提出了一种改进的 自适应特征值分解时延估计算法,该方法通过改进初值设定方法,有效改善了对时延变化的估计。 另外,通过引入一个基于相关运算的语音检测算法,提高了定位系统的抗噪声能力。实验表明在 真实的声场环境下该算法能够对单个声源的三维空间位置进行实时的定位和跟踪,系统在115m 范围内对声源的定位误差小于8cm,声源位置变化时,系统也能准确跟踪声源的位置。 关键词　麦克风阵列;声源定位;声源跟踪;AE DA算法;LMS算法 中图分类号　TP391 0　引　言 基于麦克风阵列的声源定位是声学信号处理领域中的一个重要问题,在视频会议、智能机器人、鲁棒语音识别等领域有着广泛的应用。近年来,在真实声场环境下抗混响的声源定位算法研究成为研究热点。 声源定位大致分为3类方法。第1类是基于波束成型的方法,该方法可以对单声源进行定位[1,2],也可以对多声源进行定位[3],但存在对初值敏感的问题。另外还需要知道声源和噪声的先验知识,该方法存在计算量大,不利于实时处理等缺点。第2类是基于高分辨率谱估计的方法。该方法在理论上可以对声源的方向进行有效的估计,并且适用于多个声源的情况[4]。但由于该算法是针对窄带信号,因此如要获得较理想的精度就要付出很大的计算量代价。另外这些算法无法处理高度相关的信号,因此混响会给算法的定位精度带来较大影响。第3类方法是基于时延估计的方法。该类算法计算量小,易于实时实现,近年来得到了高度重视。 基于时延估计的算法分为2个部分。第1部分为时延估计,即计算声源到两两麦克风之间的时间差;第2部分为方位估计,即根据时延和麦克风阵列的几何位置估计出声源的位置,其中时延估计最为关键。互相关法是最常用的一种时延估计算法,但是它在混响较大的情况下性能下降很多。1982年,D1H.Y oun等[5]提出了最小均方(LMS,Least Mean Square)时延估计算法,其性能和互相关法基本相当。布朗大学于1995年实现了一个实时声源定位系统[6],该系统采用相位变换的时延估计算法和线性插值方位估计算法,在混响较小的情况下能够准确的估计时延但在混响较大情况下误差较大。1997年新泽西州立大学采用相位变换法作为时 　1)国家自然科学基金(60305004)中国博士后科学基金(2003033081)资助项目 收稿日期:2004208223;修回日期:2004211211 908

食物海鲜英文翻译

水果类（fruits）： 火龙果pitaya 西红柿tomato 菠萝pineapple 西瓜watermelon 香蕉banana 柚子shaddock （pomelo）橙子orange 苹果apple 柠檬lemon 樱桃cherry 桃子peach 梨pear 枣Chinese date （去核枣pitted date ）椰子coconut 草莓strawberry 树莓raspberry 蓝莓blueberry 黑莓blackberry 葡萄grape 甘蔗sugar cane 芒果mango 木瓜pawpaw或者papaya 杏子apricot 油桃nectarine 柿子persimmon 石榴pomegranate 榴莲jackfruit 槟榔果areca nut （西班牙产苦橙）bitter orange 猕猴桃kiwi fruit or Chinese gooseberry 金橘cumquat 蟠桃flat peach 荔枝litchi 青梅greengage 山楂果haw 水蜜桃honey peach 香瓜，甜瓜musk melon 李子plum 杨梅waxberry red bayberry 桂圆longan 沙果crab apple 杨桃starfruit 枇杷loquat 柑橘tangerine 莲雾wax-apple 番石榴guava 肉、蔬菜类（livestock家畜）： 南瓜（倭瓜）pumpkin cushaw 甜玉米Sweet corn 牛肉beef 猪肉pork 羊肉mutton 羔羊肉lamb 鸡肉chicken 生菜莴苣lettuce 白菜Chinese cabbage （celery cabbage）（甘蓝）卷心菜cabbage 萝卜radish 胡萝卜carrot 韭菜leek 木耳agarics 豌豆pea 马铃薯（土豆）potato 黄瓜cucumber 苦瓜balsam pear 秋葵okra 洋葱onion 芹菜celery 芹菜杆celery sticks 地瓜sweet potato 蘑菇mushroom 橄榄olive 菠菜spinach 冬瓜（Chinese）wax gourd 莲藕lotus root 紫菜laver 油菜cole rape 茄子eggplant 香菜caraway 枇杷loquat 青椒green pepper 四季豆青刀豆garden bean 银耳silvery fungi 腱子肉tendon 肘子pork joint 茴香fennel（茴香油fennel oil 药用）鲤鱼carp 咸猪肉bacon 金针蘑needle mushroom 扁豆lentil 槟榔areca 牛蒡great burdock 水萝卜summer radish 竹笋bamboo shoot 艾蒿Chinese mugwort 绿豆mung bean 毛豆green soy bean 瘦肉lean meat 肥肉speck 黄花菜day lily （day lily bud）豆芽菜bean sprout 丝瓜towel gourd (注：在美国丝瓜或用来做丝瓜茎loofah洗澡的，不是食用的) 海鲜类（sea food）： 虾仁Peeled Prawns 龙虾lobster 小龙虾crayfish（退缩者）蟹crab 蟹足crab claws 小虾（虾米）shrimp 对虾、大虾prawn （烤）鱿鱼（toast）squid 海参sea cucumber 扇贝scallop 鲍鱼sea-ear abalone 小贝肉cockles 牡蛎oyster 鱼鳞scale 海蜇jellyfish鳖海龟turtle 蚬蛤

声源定位测试系统的制作方法

本技术公开了声源定位测试系统,包括电脑控制软件平台、控制器、功率放大器、扬声器、声音采集器，电脑控制软件平台和控制器通过USB数据线相连；所述控制器和功率放大器通过控制器对放大器数据线相连；所述功率放大器和扬声器通过放大器对扬声器数据线相连；所述声音采集器和控制器通过信号采集器对控制器数据线相连；它通过声源定位测试系统在整个输出过程中对声音的大小、方向，以及声音的种类和发出声音的声道和通道数量进行控制，来便于对具有声源定位技术的产品进行不同阶段和方式的技术测试，从而使声源定位技术测试更便捷、更准确。 技术要求 1.声源定位测试系统,包括电脑控制软件平台(1)、控制器(2)、功率放大器(3)、扬声器(4)、声音采集器(5)，其特征在于：所述电脑控制软件平台(1)和控制器(2)通过USB数据线(6)相连；所述控制器(2)和功率放大器(3)通过控制器对放大器数据线(8)相连；所述功率放大器(3)和扬声器(4)通过放大器对扬声器数据线(9)相连；所述声音采集器(5)和控制器(2)通过信号采集器对控制器数据线(7)相连；

当系统在声音输出状态时，先由电脑控制软件平台(1)发出的单个或多个声音控制指令转换成数字信号组通过USB数据线(6)传递至控制器(2)；再由控制器(2)对数字信号组进行分析处理和分流排序，并将分流排序的数字信号组采用单独、合并、部分叠加等不同的方式转换成新的排序的单个或多个模拟信号，且通过控制器对放大器数据线(8)分别传递给功率放大器(3)；后由功率放大器(3)将新的排序的单个或多个模拟信号进行放大且通过放大器对扬声器数据线(9)分别对应传递给扬声器(4)，最后由扬声器(4)将模拟信号分别转换成声信号并对外输出； 当系统在声音输入状态时，先由声音采集器(5)将所采集到的声音信号通过控制器数据线(7)传递给控制器(2)，然后由控制器(2)对声音采集器(5)所输入的模拟信号转化为数字信号，控制器(2)对数字信号进行分析处理并将处理过后的数字信号通过USB数据线(6)传递至电脑控制软件平台(1)，由电脑控制软件平台(1)将数字信号转换成图文数据显示。 2.根据权利要求1所述的声源定位测试系统，其特征在于：所述扬声器(4)为一台或多台。 3.根据权利要求1所述的声源定位测试系统，其特征在于：所述控制器对放大器数据线(8)为一根或多根。 4.根据权利要求1所述的声源定位测试系统，其特征在于：所述放大器对扬声器数据线(9)为一根或多根。 技术说明书 声源定位测试系统 技术领域 本技术涉及声学领域，具体涉及一种在一定的空间环境下，通过在不同的方位提供不同方式的声源来形成声源定位测试场所的体系。 背景技术

近场声源定位算法研究

Word文档可进行编辑 近场声源定位算法研究 近场声源定位算法研究 引言 近年来,基于麦论文联盟克风阵列得声源定位技术快速进展,同时在多媒体系统,移动机器人,视频会议系统等方面有广泛得应用.例如,在军事方面,声源定位技术能够为雷达提供一个非常好得补充,不需要发射信号,仅靠接收信号就能够推断目标得位置,因此,在定位得过程中就可不能受到干扰和攻击.在视频会议中,讲话人跟踪可为主意拾取和摄像机转向操纵提供位置信息,使传播得图像和声音更清楚.声源定位技术因为其诸多优点以及在应用上得广泛前景成为了一个研究热点.

现有得声源定位方法要紧分为三类：基于时延可能得定位方法、基于波束形成得定位方法和基于高分辨率空间谱可能得定位方法.基于时延可能得定位方法[１]要紧步骤是先进行时刻差可能,也确实是先计算声源分不到达两个麦克风得时刻差,然后依照那个时刻差和麦克风阵列得几何结构可能出声源得位置.该类方法得优点是计算量较小,容易实时实现,在单声源定位系统中差不多得到广泛应用.基于波束形成得定位方法[２]不需要直截了当计算时刻差,而是通过对目标函数得优化直截了当实现声源定位.但由于实际得应用环境中,目标函数往往存在多个极值点,因此如何优化复杂峰值得搜索过程就成为了一个重点.基于高分辨率得空间谱可能得声源定位算法,例如宽带得ｍｕｓｉｃ（ｍｕｌｔｉｐｌｅｓｉｇｎａｌｃｌａｓｓｉｆｉｃａｔｉｏｎ）方法[３]和最大似然方法[４],因其能够同时定位多个声源同时具有比较高得空间分辨率,受到了广泛得关注.

空间谱可能得方法源于阵列信号处理,其中得多重信号分类（ｍｕｓｉｃ）算法在特定条件下具有非常高得可能精度和分辨力,从而吸引了大量得学者对其进行深入得分析与研究.WwwcOm但与阵列信号处理不同得是,在声源定位中,声源在大多数情况下是位于声源近场得.为了解决这一近场咨询题,许多学者针对传统得信号模型提出了改进算法,ａｓａｎｏ等人将传统时域得ｍｕｓｉｃ[５,６]算法应用在频域中,提出了一种基于子空间得近场声源算法[７].下面来看一下近场得声源信号模型. １近场声源信号模型 传统得阵列信号处理大多是基于远场模型得平面波信号得假设,然而在声源定位得实际应用中,有非常多情况是处于声源近场得[８],例如视频会议,机器人仿真等.同时又由于麦克风阵列阵元拾音范围有限,更多得情况下定位也处于近场范围内,如今信源到达各麦克风阵元得信

各种饮食英文翻译

中式早點： 烧饼Clay oven rolls 油条Fried bread stick 韭菜盒Fried leek dumplings 水饺Boiled dumplings 蒸饺Steamed dumplings 馒头Steamed buns 割包Steamed sandwich 饭团Rice and vegetable roll 蛋饼Egg cakes 皮蛋100-year egg 咸鸭蛋Salted duck egg 豆浆Soybean milk 饭类： 稀饭Rice porridge 白饭Plain white rice 油饭 Glutinous oil rice 糯米饭Glutinous rice 卤肉饭Braised pork rice 蛋炒饭Fried rice with egg 地瓜粥Sweet potato congee 面类： 馄饨面 Wonton & noodles 刀削面 Sliced noodles 麻辣面Spicy hot noodles 麻酱面Sesame paste noodles 鴨肉面 Duck with noodles 鱔魚面 Eel noodles 乌龙面Seafood noodles 榨菜肉丝面Pork , pickled mustard green noodles 牡蛎细面Oyster thin noodles 板条Flat noodles 米粉 Rice noodles 炒米粉Fried rice noodles 冬粉Green bean noodle 汤类： 鱼丸汤Fish ball soup 貢丸汤Meat ball soup 蛋花汤Egg & vegetable soup 蛤蜊汤Clams soup 牡蛎汤Oyster soup 紫菜汤Seaweed soup 酸辣汤Sweet & sour so up 馄饨汤Wonton soup 猪肠汤Pork intestine soup 肉羹汤Pork thick soup 鱿鱼汤 Squid soup 花枝羹Squid thick soup 甜点： 爱玉Vegetarian gelatin 糖葫芦Tomatoes on sticks 长寿桃Longevity Peaches 芝麻球Glutinous rice sesame balls 麻花 Hemp flowers 双胞胎Horse hooves

各种职位的英文翻译

各种职位的英文翻译 qa 是英文 quality assurance 的简称，中文含义是质量保证； qc 是英文 quality control 的简称，中文含义是质量控 制。 IPQC 是过程检验工程师 JQE 是品质工程师 DQA 是设计品保工程师 SQE 共货商管理工程师 Administration( 行政部分) 行政主管 File Clerk 档案管理员 行政助理 Office Manager 办公室经理 行政秘书 Receptionist 接待员 办公室文员 Secretary 秘书 Inventory Control Analyst 存货控制分析 Staff Assistant 助理 Mail Room Supervisor 信件中心管理员 Stenographer 速记员 Order Entry Clerk 订单输入文员 Telephone Operator 电话操作 员 Shipping/Receiving Expediter 收发督导员 Ticket Agent 票务代理 Vice-President of Administration 行政副总裁 Typist 打字员 Executive and Managerial( 管理部分 ) Retail Store Manager 零售店经理 Food Service Manager 食品服务经理 Executive Marketing Director 市场行政总监 HMO Administrator 医疗保险管理 Assistant Store Manager 商店经理助理 Operations Manager 操作经理 Assistant Vice-President 副总裁助理 Production Manager 生产经理 Chief Executive Officer(CEO) 首席执行官 Property Manager 房地产经理 Chief Operations Officer(COO) 首席运营官 Branch Manager 部门经理 Controller(International) 国际监管 Claims Examiner 主考官 Director of Operations 运营总监 Controller(General) 管理员 Field Assurance Coordinator 土地担保协调员 General Manager 总经理 Management Consultant 管理顾问 District Manager 市区经理 Hospital Administrator 医院管理 President 总统 Import/Export Manager 进出口经理 Product Manager 产品经理 Insurance Claims Controller 保险认领管理员 Program Manager 程序管理经理 Insurance Coordinator 保险协调员 Project Manager 项目经理 Inventory Control Manager 库存管理经理 Regional Manager 区域经理 Manager(Non-Profit and Charities) 非盈利性慈善机构管理 Service Manager 服务经理 Manufacturing Manager 制造业经理 Vending Manager 售买经理 Telecommunications Manager 电信业经理 Vice-President 副总裁 Transportation Manager 运输经理 Warehouse Manager 仓库经理 Education and Library Science( 教育部分 ) Daycare Worker 保育员 ESL Teacher 第二外语教师 Developmental Educator 发展教育家 Head Teacher 高级教师 Foreign Language Teacher 外语教师 Librarian 图书管理员 Guidance Counselor 指导顾问 Music Teacher 音乐教师 Library Technician 图书管理员 Nanny 保姆 Physical Education Teacher 物理教师 Principal 校长 School Psychologist 心理咨询教师 Teacher 教师 Special Needs Educator 特种教育家 Teacher Aide 助理教师 Art Instructor 艺术教师 Computer Teacher 计算机教师 College Professor 大学教授 Coach 教练员 Assistant Dean of Students 助理训导长 Archivist 案卷保管员 Vocational Counselor 职业顾问 Tutor 家教、辅导教师 Auditor 审计师 Accountant 会计员，会计师 Administration Assistant 行政助理 Administrator 行政主管 Assistant Manager 副经理 Assistant Production Manager 副厂长 Business Manager 业务经理 Cashier 出纳员 Chief Accountant 总会计主任 Chief Engineer 总工程师 Civil Engineer 土木工程师 Clerk 文员(文书) Director 董事 Electrical Engineer 电气工程师 Executive Director 行政董事 Executive Secretary 行政秘书 Financial Controller 财务总监 Foreman 领班，组长 General manager 总经理 Junior clerk 低级文员（低级职员） Manager 经理 Marketing Executive 市场部主任 Marketing Manager 市场部经理 Marketing Officer 市场部办公室主任 Mechanical Engineer 机械工程师 Merchandiser 买手（商人） Messenger 信差（邮递员） Office Assistant 写字楼助理（办事员） Administrative Director Executive Assistant Executive Secretary General Office Clerk

食物的英文翻译

食物的英文翻译 [日期：2015-03-10] 来源：作者：[字体：大中小] 水果类西红柿tomato；菠萝pineapple；西瓜watermelon；香蕉banana；柚子shadd ock（pomelo）；橙子orange；苹果apple；柠檬lemon；樱桃cherry；桃子peach；梨p ear；枣Chinese date（去核枣pitted date）；椰子coconut；草莓strawberry；树莓ras pberry；蓝莓blueberry；黑莓blackberry；葡萄grape；甘蔗sugar cane；芒果mango；木瓜pawpaw或者papaya；杏子apricot；油桃nectarine；柿子persimmon；石榴pomeg ranate；榴莲jackfruit；槟榔果areca nut；西班牙产苦橙bitter orange；猕猴桃kiwi fruit or Chinese gooseberry；金橘cumquat；蟠桃flat peach；荔枝litchi；青梅greengage；山楂果haw；水蜜桃honey peach；香瓜、甜瓜musk melon；李子plum；杨梅waxberr y red bayberry；桂圆longan；沙果crab apple；杨桃starfruit；枇杷loquat；柑橘tange rine；莲雾wax-apple；番石榴guava 肉、蔬菜类南瓜（倭瓜）pumpkin cushaw；甜玉米Sweet corn；牛肉beef；猪肉p ork；羊肉mutton；羔羊肉lamb；鸡肉chicken；生菜、莴苣lettuce；白菜Chinese cabb age（celery cabbage）；甘蓝、卷心菜cabbage；萝卜radish；胡萝卜carrot；韭菜leek；木耳agarics；豌豆pea；马铃薯（土豆）potato；黄瓜cucumber；苦瓜balsam pear；秋葵okra；洋葱onion；芹菜celery；芹菜杆celery sticks；地瓜sweet potato；蘑菇mush room；橄榄olive；菠菜spinach；冬瓜（Chinese）wax gourd；莲藕lotus root；紫菜la ver；油菜cole rape；茄子eggplant；香菜caraway；枇杷loquat；青椒green pepper；四季豆、青刀豆garden bean；银耳silvery fungi；腱子肉tendon；肘子pork joint；茴香fennel（茴香油fennel oil 药用）；鲤鱼carp；咸猪肉bacon；金针蘑needle mushroom；扁豆lentil；槟榔areca；牛蒡great burdock；水萝卜summer radish；竹笋bamboo sho ot；艾蒿Chinese mugwort；绿豆mung bean；毛豆green soy bean；瘦肉lean meat；肥肉speck；黄花菜day lily（day lily bud）；豆芽菜bean sprout；丝瓜towel gourd (注：在美国丝瓜或用来做丝瓜茎loofah洗澡的，不是食用的) 海鲜类虾仁Peeled Prawns；龙虾lobster；小龙虾crayfish；蟹crab；蟹足crab claw s；小虾shrimp；对虾、大虾prawn；（烤）鱿鱼（toast）squid；海参sea cucumber；扇贝scallop；鲍鱼sea-ear abalone；小贝肉cockles；牡蛎oyster；鱼鳞scale；海蜇jellyfi sh；鳖、海龟turtle；蚬、蛤clam；鲅鱼culter；鲳鱼butterfish；虾籽shrimp egg；鲢鱼、银鲤鱼chub silver carp；黄花鱼yellow croaker 调料类醋vinegar 酱油soy 盐salt 加碘盐iodized salt 糖sugar 白糖refined sugar 酱soy sauce 沙拉salad 辣椒hot（red）pepper 胡椒（black）pepper 花椒wild peppe r Chinese prickly ash powder 色拉油salad oil 调料fixing sauce seasoning 砂糖gran ulated sugar 红糖brown sugar 冰糖Rock Sugar 芝麻Sesame 芝麻酱Sesame paste 芝麻油Sesame oil 咖喱粉curry 番茄酱（汁）ketchup redeye 辣根horseradish 葱shal lot （Spring onions）姜ginger 蒜garlic 料酒cooking wine 蚝油oyster sauce 枸杞（枇

各种职位的英文翻译

各种职位的英文翻译 qa是英文quality assurance 的简称，中文含义是质量保证；qc是英文quality control的简称，中文含义是质量控制。IP QC是过程检验工程师 JQE是品质工程师 DQA是设计品保工程师 SQE供货商管理工程师 Administration(行政部分） Administrative Director 行政主管File Clerk 档案管理员 E xecutive Assistant 行政助理O ffice Manager 办公室经理 E xecutive Secretary 行政秘书Receptionist 接待员 General Office Clerk 办公室文员Secretary 秘书 Inventory Control Analyst 存货控制分析Staff Assi s tant 助理 Mail Room Supervisor 信件中心管理员Stenographer 速记员 Order E ntry Clerk 订单输入文员Telephone Operator 电话操作员 Shipping/Receiving E xpediter 收发督导员Ticket Agent 票务代理 Vice-P resident of Administration 行政副总裁Typi s t 打字员 E xecutive and Managerial(管理部分) Retail Store Manager 零售店经理Food Service Manager 食品服务经理 E xecutive Marketing Director 市场行政总监HMO Administrator 医疗保险管理 Assi s tant Store Manager 商店经理助理Operations Manager 操作经理 Assi s tant Vice-P resident 副总裁助理P roduction Manager 生产经理 Chief E xecutive O fficer(CE O) 首席执行官P roperty Manager 房地产经理 Chief Operations O fficer(COO) 首席运营官Branch Manager 部门经理 Controller(International) 国际监管Claims E xaminer 主考官 Director of Operations 运营总监Controller(General) 管理员 Field Assurance Coordinator 土地担保协调员General Manager 总经理 Management Consultant 管理顾问District Manager 市区经理 Hospital Administrator 医院管理P resident 总统 Import/E xport Manager 进出口经理P roduct Manager 产品经理 Insurance Claims Controller 保险认领管理员P rogram Manager 程序管理经理 Insurance Coordinator 保险协调员P roject Manager 项目经理 Inventory Control Manager 库存管理经理Regional Manager 区域经理 Manager(Non-P rofit and Charities) 非盈利性慈善机构管理Service Manager 服务经理 Manufacturing Manager 制造业经理Vending Manager 售买经理 Telecommunications Manager 电信业经理 Vice-P resident 副总裁 Transportation Manager 运输经理Warehouse Manager 仓库经理 E ducation and Library S cience(教育部分) Daycare Worker 保育员E SL Teacher 第二外语教师 Developmental E ducator 发展教育家Head Teacher 高级教师 Foreign Language Teacher 外语教师Librarian 图书管理员 Guidance Counselor 指导顾问Music Teacher 音乐教师 Library Technician 图书管理员Nanny 保姆 P hysical E ducation Teacher 物理教师P rincipal 校长 School P sychologist 心理咨询教师Teacher 教师 Special Needs E ducator 特种教育家Teacher Aide 助理教师 Art Instructor 艺术教师Computer Teacher 计算机教师 College P rofessor 大学教授Coach 教练员 Assi s tant Dean of Students 助理训导长Archi v ist 案卷保管员 Vocational Counselor 职业顾问Tutor 家教、辅导教师 Auditor 审计师 Accountant 会计员，会计师 Administration Assistant 行政助理 Administrator 行政主管 Assi s tant Manager 副经理 Assi s tant P roduction Manager 副厂长 Business Manager 业务经理 Cashier 出纳员 Chief Accountant 总会计主任 Chief E ngineer 总工程师 Civil E ngineer 土木工程师 Clerk 文员（文书） Director 董事 E lectrical E ngineer 电气工程师