[考研英语]英语翻译及阅读资料-经济学人6.pdf

《[考研英语]英语翻译及阅读资料-经济学人6.pdf》由会员分享，可在线阅读，更多相关《[考研英语]英语翻译及阅读资料-经济学人6.pdf（34页珍藏版）》请在淘文阁 - 分享文档赚钱的网站上搜索。

1、Technology Monitor科技前沿Correct me if Im wrong.有错请纠正Better speech-recognition technology语音识别技术的新进展Jul 12th 2010THERE is often something sweet,intimate even,about couples who finish each others sentences.But it can also be a source of irritation,especially when they get it wrong.A similar irritation（mi

2、nus the sweetness）is often felt by users of speech-recognition software,which still manages togarble and twist even the most clearly spoken words.Perhaps the solution lies in a more intimateexchange between user and software.情侣间品味对方未说完的话语时经常有种甜丝丝的感觉，甚至会感到亲切。但这种半截话也可能惹恼对方，尤其是产生了错误解读的情况下更是如此。语音识别软件仍然在

3、千方百计地窜改和歪曲甚至是用户最清晰的吐词，这同样会使人感到恼火（用户对软件当然不会有什么甜蜜的感觉）。解决这个问题的办法或许就只能是用户和软件进行更加密切的交流。Modem speech-recognition programs do not merely try to identify individual words as they arespoken;rather,they attempt to match whole chunks of speech with statistical models of phrasesand sentences.The rationale is th

4、at by knowing statistical rules of thumb fbr the way in whichwords are usually put together-an abstract probabilistic approximation of grammar,if youwillit is possible to narrow the search when attempting to identify individual words.Forexample,a noun-phrase will typically consist of a noun preceded

5、 by a modifier,such as an articleand possibly also an adjective.So if part of a speech pattern sounds like“ball，the odds of itactually being ball will increase if the utterances preceding it sound like the and bouncy”.现代语音识别程序不只是要识别单词，相反，它们通过统计用户使用句子和短语的模式试图能够识别成段的话语。其基本原理是通过运用统计学上的经验公式归纳出单词组合的方式，这是

6、一个抽象的运用在语法学上的概率型逼近问题，当软件需要识别用户吐出的单词时.，运用这种方法可以缩小搜索范围。当然这一切需要用户的配合。例如，名词短语通常由一个名词和前面的一个冠词构成，如一篇文章（an article）等，也可以是由一个名词加一个形容词组成。因此，如果一个声谱上语音的一部分听起来像是“ball”（球），而在这个发音前面有听起来像是定冠词“the”和“bouncy”（跳跃的）这样的声音，这个发音确实是“球”的可能性就增加了。Although this so-called continuous speech-recognition approach has indeed improv

7、ed accuracy,itis by no means infallible.Moreover,when it gets things wrong,it often does so spectacularly.Theproblem is that,as a direct consequence of this technique,the misidentification of even a singleword can take the program off on a completely different path as it tries to predict what the re

8、st ofthe sentence is likely to be.这种所谓的连续语音识别方法虽然确实能够提高语音识别的正确率，但还做不到万无一失。此外，它一旦识别错误，展现在面前的文字往往令人啼笑皆非。采用这种语音识别技术，只要有一个单词被识别错了，程序就会根据这个单词去猜测句子的剩余部分，因而就会沿着一条完全错误的路径走下去，结果也就可想而知了。Though such errors are inevitable,there may be a way to let speech-recognition programs take thepain out of making correctio

9、ns.Per Ola Kristensson and Keith Vertancn,at the University ofCambridges Computer L aboratory,have developed a method of allowing speech-recognitionprograms to share their thoughts,as it were,with the user,in order to speed up the correctionprocess.Their solution,called Parakeet,is a touch-screen-ba

10、sed interface for phones and othermobile devices,which not only displays the words,phrases or sentences that scored highest in theprograms statistical model,but also any close contenders.This allows the user to selectalternatives easily,with a quick tap of the finger.More subtly,if none of the predi

11、cted sentences isentirely correct,yet collectively they contain the words that were spoken,the user can simply slidehis finger across the appropriate words to link them up.虽然这样的错误无法避免，但要让语音识别程序去改错，则难度太大。现在有种方法可能有助于解决这一难题。剑桥大学计算机实验室的珀欧莱克里斯特森(Per Ola Kristensson)和基思维特尼(Keith Vertanen)研究出-种方法，这种方法在以往版本

12、的软件上已经采用过,就是让语音识别程序和用户进行人机对话以加强了解，使改错的进程得以加快。他们用这种方法开发的软件被称作“长尾鹦鹉”，使用了用于手机和其他移动通讯设备的触屏界面，显示在屏幕上的不仅有程序根据统计模型筛选出的单词、短语或句子，而且还包括了所有与它们近似的成份。这使得用户选择起来更加方便，手指轻轻一点，需要的词就选出来了。更精妙的是，如果程序先前猜测的句子不正确，而这些句子包含了用户所吐出的单词，用户就只需用手指在屏幕上滑过适当的单词将它们排列起来就可以了。In a sense,all Parakeet is doing is allowing the user to see wh

13、ich alternative words or sentencesthe program would have predicted.The difference is that existing programs require the user tocorrect each word individually,from a drop-down list of alternatives,or else to retype or reutterthe words.What is frustrating about this,says Dr Kristensson,is that more of

14、ten than not thecorrect strings of words were recognised,but rejected by the speech-recognition program onstatistical grounds.Parakeet makes them all available to the usei;在某种意义上，“长尾鹦鹉 就是将程序预测的备选单词与句子直观地展示在用户面前。过去类似的程序需要用户从 系列下拉菜单中选词来逐个纠正每个单词，要么就是重新打字或重新再说一遍。克里斯特森说，很多情况下是正确的单词串被识别出来，却被语音识别程序基于统计规则而舍

15、弃了，这才令人沮丧。而“长尾鹦鹉”把所有的备选项都列出来由用户来做出选择。The prototype uses an open-source speech-recognition program called Pocket Sphinx,developedat Carnegie Mellon University,in Pittsburgh,but Dr Kristensson reckons it would be easy to applythe same approach to commercially available programs like Nuances Dragon.So

16、far DrKristensson and Dr Vcrtanen have carried out only limited trials on a handful of people.Even so,these have achieved operating rates of around 22 words per minute-considerably higher than the16 an average user can achieve using predictive texting.With the likes of Google,Nuance andVlingo now of

17、fering mobile speech-recognition services for phones,and the development ofentertainment systems and vehicle communication,such as Fords Sync platform,Parakeet may beflying into a growing market.匹兹堡的卡耐基梅隆大学曾开发了一种叫做“袖珍斯芬克斯”的开源语音识别软件，“长尾鹦鹉“就是在此基础上 进行研制的。克里斯特森博士估计将这种方法用于如纽安斯公司的“龙”这类商业软件的开发上并不难。迄今为止，克里斯

18、特森和维特尼两位博士仅仅用十几个人对这套软件进行了有限的试验。但即使如此，这些试验还是取得了每分钟22 个单词的识别速度，而以往采用预测下文的方法用户平均只能达到每分钟16个单词的输出速度，如此对比就可看出使用 长尾鹦鹉 后文本输出的速度要快得多。随着谷歌、纽安斯和Vlingo等公司纷纷给手机增加了语音识别功能，以及如福特的SYNC这样的车载多媒体通讯娱乐技术的发展，语音识别技术的市场前景无限广阔，“长尾鹦鹉 必将飞得更高。The controversies in climate science气候学界的争议Science behind closed doors围城里的科学Two new re

19、ports say the science of climate change is fine,but that some scientists and theinstitutions they work in need to change their attitudes最新的两份报告表明气候学界的学术气氛仍然是良好的，但是某些科学家以及他们所工作的机构必须改变对待公众的态度。Jul 8th 2010THE winter of 2009 was a rough time fbr climate science.In November,in the run-up to theCopenhagen

20、 climate conference,over 1,000 private e-mails from and to researchers at theClimatic Research Unit(CRU),a part of the University of East Anglia(UEA)in Britain,appearedon the internet,presumably after being stolen.At the same time a controversy was bubbling up inIndia over a claim in the 2007 assess

21、ment report by the Intergovernmental Panel on ClimateChange(IPCC)that the Himalayas could lose all their glaciers in 25 years,which was wrong.These events seemed to provide evidence of embarrassing incompetence,at the very least.2009年冬季对气候学界来说是段难熬的日子。当 年 11月份，正当哥本哈根气候峰会即将召开之际，英国东安格利亚大学(University of

22、 East A nglia)气候研究部门(C R U)的科学家之间超过1000封私人邮件被盗窃泄露到互联网上。与此同时，有关联合国政府间气候变化专门委员会在2007递交的一份评估报告中提出的喜马拉雅山脉的冰川将于25年之内全部消失的错误结论的争议也开始在印度浮现出来。不管怎么说，这些事件反映出了气候学界令人尴尬的无能。Explanations were demanded and committees were fbnncd to deliver them.This week two ofthose committees reported.For the CRU and what became

23、known as climategate”,anindependent panel was created by UEA and chaired by Muir Russell,a fbnner vice-chancellor ofthe University of Glasgow.The Dutch environmental-assessment agency was asked to look forother errors in the regional analyses of the IPCCs report.Both the reports conclude that thesci

24、ence of climate is sound and that the professional characters of the scientists involved areunimpeached.But they raise important issues about how to do science in such an argumentativearea and under new levels of scrutiny,especially from a largely hostile and sometimes expertblogosphere.专门的组委会被组建起来以

25、调查真相，本周其中的两家组委会做出了相关的报告。一个由东安格利亚大学组织并山格拉斯哥大学(University of Glasgow)前任副校长Muir Russell主持的独立委员会(下文称罗素委员会)被组建起来调查该气候研究部门事件-此事后来被人们称为“气候门”。荷兰的环境评估部门被指定调查联合国政府间气候变化专门委员会的报告中有关区域气候分析的其他组误。来自双方的报告都得出结论说气候学界的学术气氛是科学健全的，被牵连的科学家的职业操守也是无可厚非的。但这些事件也让人们意识到一些重要的问题，就是在个如此具有争议性的领域，尤其是面对来自基本上都是心怀敌意并具有定专业水准的大众博客圈的仔细审查

26、，科学家应该怎样做科学研究。The Dutch agency found a few errors in the relevant chapters of the IPCCs report,though noneamounted to much.It also raised questions about concentrating on bad or worst-case possibilitiesrather than a range of outcomes.The agency did not say this was a bad thing-policymakers needto ha

27、ve the most critical information flagged upbut it thinks it would be better to explain moreclearly what is going on.荷兰的环境部门在联合国政府间气候变化专门委员会的报告中相关章节找到了 一些纸误,但是都无关大碍。这也同样引起了另外一个问题，我们是应该关注那些可能出现的最糟糕的情况还是一系列可能的结果？调查部门没有说宣传最糟糕的可能性有什么不对，因为决策者需要科学家向他们强调最关键的信息，但是如果能够向大众解释清楚到底是怎么回事就最好了。Martin Parry,who in 20

28、07 was co-chair of the relevant IPCC working group,says there was not aconscious decision to highlight negative effects,but to highlight important ones,as measured bysuch things as scale and irreversibility.The important effects are negative ones:this is why peopleare worried about climate change.A

29、tendency fbr the IPCC process to produce outputs moreworrying,at the margins,than its inputs does not necessarily show bias.It may reflect accurateexpert assessment.But the risk that it is a sort of selfreinfbrcing groupthink merits attention.2007年联合国政府间气候变化专门委员会相关工作小组的联合主席Martin Parry说，科学家们潜意识中并没有去

30、强调气候变化的负面影响，而是去强调以影响的广度和可逆性来衡量最重要的结果。而其实最重要的是前者，这也是公众关心气候变化的原因。联合国政府间气候变化专门委员会的报告倾向于得出比实际情况更加严重的结论并不一定意味着其报告与实际情况存在偏差。它也同样可能反应正确的评估结果。不过当然，这些结论是否仅仅是由群人互相危言耸听造成的也同样值得人们关注。Open to criticism欢迎批评A fonn of groupthink certainly seems to have been at work in the climategate e-mails.The Russellcommittee was

31、 most exercised by a lack of openness at the CRU,in part explained,but notexcused,by a sort of a siege mentality.The committee found that the scientists committed nothingclose to fraud.It showed that the data needed to reconstruct CRUs temperature records werewidely available.Informed by a warts-and

32、-all account of peer review from Richard Horton,editorof The L ancet,a medical journal,the committee took the researchers9 harsh behaviour towardscritics and leniency towards allies not as unduly partial or aggressive,but as part of the rough andtumble of interaction in an area of science that has b

33、ecome heavily contested.As an eminenthistorian of science put it to a naive scientist when the story first broke,Everything you believeto be true once looked like this.”在“气候门”邮件中显然存在着一种集体性思维。罗素委员会在东安格利亚大学气候研究部门的调查本身就缺乏透明度，这部分可以从某种受困心态来解释，但绝不能以此为借口。委员会的调查表明相关科学家没有任何学术造假的迹象，重建该部门的气温记录所需的原始数据也随处可得。受医学杂

34、志 柳叶刀(The L ancet)主编Richard Horton关于同行评审瑕瑜共存的解释所启发，委员会没有将相关研究人员对批评者的刻薄行为和对支持者的过分宽容视为是不当的偏见或挑衅，而是视为在一个被广泛质疑的科学领域中研究人员与公众之间艰难而匮乏的沟通的部分体现。就像位著名的科学史学家在事件刚发生时将之归咎于科学家天真的想法时所说的那样，“你现在所信以为真的一切科学理论都曾有过这么一段不为大众接受的阶段。”But the committee did criticise the researchers for an unwillingness to pass on data to thei

35、r critics,for failing to specify which weather stations they were using,for keeping quiet IPCC discussionsthat should have been public,and so on.This flowed through into“clear incitemenf,to delete filesrather than have them surrendered under Britains freedom-of-information act.The committeefound UEA

36、s procedures on freedom of information poor.同时委员会也确实批评了相关研究人员的一些不当行为，比如不情愿将数据交给批评者，没有注明他们的数据来源是哪座气象站，以及在本该公开的联合国政府间气候变化专门委员会的讨论中保持神秘等。当他们选择删除对己不利的文件而不是在英国信息公开法案的要求下交出这些数据的时候，整个事件演变成为对大众神经的一次公然的刺激。委员会发现东安格利亚大学信息公开的措施相当不力。Rather remarkably,neither the Russell committee or the university has asked Phil

37、 Jones,who ranthe CRU,whether he actually deleted e-mails with the intention of foiling subsequent requestsunder the act.The university says it takes very seriously the need to improve its openness.At thesame time it has appointed Dr Jones to a new position as director of research at theCRUdefinitel

38、y not a demotionwhile abolishing the role of director and integrating the unitmore fully into its school of environmental sciences.异乎寻常的是，罗素委员会和东安格利亚大学都没有质问该部门主管Phil Jones删除这些邮件的目的是否是为了阻止之后在信息公开法案要求下调阅资料的请求。东安格利亚大学说该校在信息公开的要求上非常认真，撤销了气候研究部门主任一职并将促使该部门与环境科学学术界接轨，然而与此同时却以该校气候研究部门研究主任的新职务聘用Phil Jones博士

39、-这当然不是降职了。In doing this UEA accepts that Dr Joness role in one of the most famous aspects ofclimategatehis“hide the decline e-mailwas“misleading”，as the Russell report puts it,without deliberately intending to be so.The growth of some trees,as recorded in their rings,trackstemperature from the 19th

40、century to the 1960s,but then ceases to do so:the two records diverge.In a graph prepared fbr the World Meteorological Organisation in 1999,Dr Jones cut off thedivergent part of one set of tree-ring data and spliced on data from thennometers.The scientificliterature contained full discussions about

41、the problems of divergence and various ways of dealingwith them,but Dr Joness chart had no readily accessible explanations or caveats.东安格利亚大学的做法表明它接受了罗素委员会报告中关于Jones博士在此次“气候门”中所扮演之角色的说法，Jones博士掩藏气候变冷证据确实具有误导性，但是没有误导大众的主观意愿。记录在年轮上的树木生长情况反映了从19世纪到20世纪60年代的气候信息，但是随后它们就不再记录了：因为两种结果背道而驰。在 1999年为世界气候组织所准备

42、的图表中，Jones博士裁掉了树木年轮数据上跟气候变暧所矛盾的部分并粘贴上从温度计测量而来的数据。科学文献中有关于此背离问题的完整讨论并且有多种处理方法，但是Jones博士的图表上没有任何显见的解释或注释。The Russell report is thorough,but it will not satisfy all the critics.Nor does it,in some ways,fulfil its remit.One of the enduring mysteries of climategate is who chose the e-mails releasedonto t

43、he internet and why they did so.These e-mails represented just 0.3%of the material on theuniversitys backup server,from which they were taken.This larger content has still not reallybeen explored.罗素委员会的报告虽然很详尽，但也仍然不能使反对者满意。在某种程度上它同样也没有完成它的职责。关于气候门一个持久的疑问是谁筛选了被泄漏到互联网上的电子邮件以及它们为什么要这么做。这些邮件仅仅是其所在的该校备份服

44、务器上所有资料的0.3%。服务器上更多的内容还没有被调查过。And then there is the science.An earlier report on climategate from the House of Commonsassumed that a subsequent probe by a panel under L ord Oxburgh,a former academic andchairman of Shell,would deal with the science.The Oxburgh report,though,sought to show onlythat th

45、e science was not fraudulent or systematically flawed,not that it was actually reliable.Andnor did Sir Muir,with this third report,think judging the science was his job.So,for verdicts as towhether the way that tree-rings from the Yamal peninsula in Siberia were treated by the CRUproduced good resul

46、ts,those following the affair will have to look for future developments injournals and elsewhere.The mode of production has been found acceptable,but the product is fbrothers to judge.Science,in the normal run of things,should do that;and if it does so in a moreopen,blogosphere-inclusive way some go

47、od will have come of the affair.然后是关于学术研究方面的问题。下议院关于气候门的份更早的报告本以为由前研究人员、现任壳牌石油公司董事的Oxburgh勋爵领导的小组会在随后的调查中处理这个问题。但是 Oxburgh勋爵的报告仅仅是致力于证明相关研究不存在学术造假或是系统性的缺陷，而不是证明其研究结论确凿可信。M uir阁下在他的报告中也同样不认为对这门学科下判断是他的职责所在。因此，如果想知道关于那些西伯利亚亚马尔半岛的树木年轮数据是否被东安格利亚大学气候研究部门篡改过以得到更好的结果的裁决，关心此事的人只好从相关学术刊物的后续进展或其他途径得到线索了。这套科学研

48、究的过程被认为是可以接受的，但是由此得到的结果则任人评价。在正常的情况下科学只能做到这样，但是如果这样的做法是在一个更开放、对社会大众更包容的环境下，那么这起事件也不至于会这么糟糕。Quantum computing量子计算A quantum hop量子跃进Not a leap,perhaps,but two important steps on the way to making quantum computing practical在实践量子计算的道路上，这也许算不上飞跃，只不过是前进了两达步。Jun 24th 2010SOME technologies seem a long time

49、coming.Their story has a familiar script:a breakthrough ishailed;overblown expectations are aroused;disappointment follows.Away from the publicsimpatient gaze,however,tinkering scientists turn out incremental improvements,furtivelycatching up with the revolutionary rhetoric of yore.This appears to h

50、ave been the case with thesequencing of the human genome.It also illustrates the recent history of quantum computing.一些科技似乎要花很长的时间，它们的故事也大同小异：人们对这些突破，先是破众星捧月、赞赏有佳，最后却无不乘性而来，败性而归。然而科学家却与浮躁盲从的大众不同，他们在这些领域缝缝补补，不断稳步前行，默默紧随昨日之辉煌成就。人类基因组测序工作便是如此，今日之量子计算革命亦然。An ordinary computer stores and processes inform