本地化翻译(Localization)：如何做好翻译
机器翻译中的本地化翻译方法研究
机器翻译中的本地化翻译方法研究机器翻译（Machine Translation, MT）是指利用计算机自动将一种语言的文本翻译成另一种语言的技术。
随着全球化的发展，跨语言交流日益频繁，机器翻译在实现不同语言间的沟通和理解方面起到了重要的作用。
然而，由于不同语言之间的差异和语言规范的个体化，机器翻译仍然面临一系列的挑战，其中之一就是如何进行本地化翻译。
本地化翻译（Localization Translation）是指根据文化和习俗的差异，对产品、服务和内容进行适应和调整的过程。
在机器翻译中，本地化翻译是指将翻译结果进行进一步的调整和优化，以符合目标语言用户的习惯和偏好。
本地化翻译的目标是使机器翻译结果更符合目标语言使用者的需求，并提高翻译的质量和可理解性。
为了进行本地化翻译，研究人员提出了一系列的方法和技术。
以下将介绍几种主要的本地化翻译方法：1. 术语本地化术语是特定领域中的专业词汇，对于保持文本的一致性和准确性至关重要。
在机器翻译中，术语本地化是指将源语言中的术语翻译成目标语言中的相应术语。
为了实现术语本地化，可以使用术语词典和术语数据库，将术语对应关系作为机器翻译的规则进行翻译。
此外，还可以利用机器学习技术，通过分析大量的术语对齐数据，自动学习术语之间的对应关系。
2. 语言风格本地化不同语言有不同的语法结构、表达方式和文化背景，使得翻译结果的风格和口语化程度存在差异。
在本地化翻译中，语言风格本地化是指将源语言文本的语言风格和表达方式调整为目标语言的习惯和偏好。
例如，英语中的表达方式可能更正式和直接，而在中文中可能更婉转和间接。
为了实现语言风格本地化，可以使用统计机器翻译中的语言模型和翻译模型，对翻译结果进行进一步的调整和优化。
3. 文化差异本地化不同文化背景和社会习俗对语言的使用和理解产生了深远影响。
在本地化翻译中，文化差异本地化是指将源语言文本中的文化细节和习俗调整为目标语言的文化背景和社会习惯，以便目标语言用户更好地理解和接受翻译结果。
本地化翻译过程
Demystifying the Costs of Localization andTranslation (1)∙I already understand localization. Or do I?∙What do translators really do with my product?∙Why is localization so expensive?Introduction―Why is localization so expensive?!‖We hear this question a lot from our clients, and at one time we had a short answer for it:Words.Our clients usually just scowled at us when we gave that answer, and so we elaborated a bit:The expense-perspective: You paid to create yourEnglish-language product, but because your engineers andwriters use English words, it looked to you as though you didn’tpay anything to create it. Now you need to write a fat check tosomebody in order to create other versions, and you’re annoyedbecause ―all they’re doing is translating,‖ which feels like child’splay compared to the work you’ve done.The revenue-perspective: Your investment in theEnglish-language product will be returned by lots and lots ofEnglish-speaking people who will give you money because yousolved their problems. Similarly, localization is an investment in a[German/Japanese/Korean/Russian/ French/...] product, and thisinvestment will be returned by lots and lots ofnon-English-speaking people who will give you money for solvingtheir problems.In other words, there is an expense-side and a revenue-side to the coin of localization.We also hear that many people in technology consider the process of delivering international products a ―black box‖. Accordingly, Part I of this paper explains the terms and steps in ―Internationalization‖ and ―Localization,‖ with a few grisly details that are second nature to translation professionals, but which look like a black box to most of their clients. In Part II we describe where high localization costs come from, and what organizations can do about the side of the black box over which they do have some control. (When all is said and done, it still comes down to the first answer: Words.)Background: InternationalizationIt is best in the long run to first internationalize the software, as in the Portuguese example in Figure 1, so that no matter how it changes for user interface or business logic, the code base at the core of the product is always the same (sometimes called a single worldwide binary). To externalize those features and characteristics that can change from one region (or locale) to another—e.g., language, color scheme and accounting standards in software, or narrative examples, conditional text and legal text in documentation—is to place them in separate resource files that vary from locale to locale, then call them from a single, common code base.Figure 1This process of internationalization (or I18n, because there are 18 letters between the ―I‖ and the ―n‖) makes life easier for a number of people along the value chain.character set enabling, overall software architecture – but ultimately, companies internationalize their products not only because it’s better engineering, but also because they can reduce time to market and make/save more money in all regions.Of course, it’s nearly impossible to get I18n right the first time, and it usually requires multiple product cycles to refine the process, so the enlightened organization takes a long, patient view. Impatient organizations may leave locale-specific elements, such as strings or error messages, inside the code base, usually to get the English-language product to market in a hurry. This is not evil, but there are costs associated with it later in the process. Companies which outsource I18n effectively leave much of the process inside the black box. For those companies whose own engineers and writers internationalize their products, I18n is not so mysterious.Background: LocalizationFew companies, however, localize their products themselves, because few companies can justify the cost of keeping translators and localization engineers on staff. Localization (or L10n) is the process of delivering a product that meets the needs of users in a specific locale, and because most companies outsource this function, it usually lands squarely in the black box. The L10n process is not completely different from or contrary to the process of creating most technology deliverables. When the product has been properly internationalized, L10n is a parallel function that takes a copy of the resources from their normal flow in development, modifies the copy for the needs of a specific locale, and joins the original further downstream in the development flow (Figure 2).Figure 2There are a few localization concepts, however, which are alien to most people who build technology products, and these concepts are inside the black box. Key ConceptsGlossary (Terminology List)To ensure uniformity of translation throughout the product (and, as the international effort grows, throughout the company), it is a good practice to put in place a glossary, which contains approved translations of key words andphrases. A translation glossary gives the equivalent of the key terms in the target language.Figure 3The Explanation column in Figure 3 is very important for preserving contextual information for the benefit of the translators. Note also that the glossary plays the important role of dictating what should not be translated.Here are some key moments in the life of a glossary:1. Client hands off early version of product to localization vendor forcreation of glossary.2. Localization vendor compiles list of key terms, with contextualcomments.3. Client conducts training session for translators and editors (optional,and too often overlooked)4. Translator translates (or, in some cases, doesn’t translate) intotarget-language equivalents.5. Vendor returns glossary draft to client.6. Client sends glossary out for review by stakeholders most likely tocomplain about undesirable translations, in order to avoid thesecomplaints once the product has been released. (This is extremelyimportant, and should be performed by in-country partners andco-workers whose livelihood depends on the quality of the translation.)7. Client returns glossary comments to vendor, who incorporates them.8. Once approved, the glossary goes to translators, reviewers, editors andclient for continued reference.A typical glossary will contain a few dozen up to a few hundred terms.Simultaneous shipment (Sim-ship)Handing off resources for translation early in development allows the localization process to begin sooner, but the original resources are more likely to change and grow, requiring additional handoffs before release of the original. On the other hand, handing off the resources after the release of the original ensures that they are frozen, but delays the launch into the regional markets (see Figure 2).Simultaneous shipment in multiple languages is attainable, but it usually takes several localization cycles, as well as ironclad buy-in from upper management. The localization process needs to be an integral part of mainstream development so that changes to the original move quickly into the localized versions.Machine Translation (MT)The urge to automate as much of human effort as possible has also touched the specialty of translation. Since the 1950’s, the field of computational linguistics has contributed a great deal to the technology behind computerized translation, and the road ahead is filled with promise.In MT, the computer applies rules and algorithms for syntax, morphology, semantics and other rules to translate text into a destination language. Another approach uses statistical models to arrive at the most likely translation for the input text. Depending on the source-destination language-pair (some are better matched than others), the unedited result will almost always preserve meaning, but it will rarely be as natural as if translated by a human native speaker.Some localization vendors use MT as adjunct technology in their translation workflow. It can save money and time as an interim step to a post-translation editing pass, but by and large, human language is too old and MT not (yet) old enough for most clients to entrust completely the localization of, say, technical documentation entirely to a computer.Translation Memory (TM) and Computer-Assisted Translation (CAT) Tools While also computer-based, these tools differ from Machine Translation. The goal of MT is for the computer to bear the brunt of the translation work, whereas TM and CAT tools help the translator do his/her own translating better, more accurately and faster. Localization vendors in the 21st century must use these tools not only to compete on price, but also to meet market expectations of consistency and quality of translation. Currently prominent products in theTM/CAT category include Catalyst, Déjà Vu, Passolo, SDLX and Trados, among many others.TM expands on the idea of the glossary. Beyond correlating a few dozen or hundred key terms, TM creates a one-to-one correlation between all of the source text and all of the destination text in the entire product and places these correlated pairs into a scalable database. Figure 4 shows source text on the left and destination text on the right.Figure 4The database may contain text from software resources, documentation, Web pages and marketing collateral, making all of it available as reference material to any translator working on any of these projects. In addition, fuzzy-matching algorithms rate approximate matches, so if the TM software finds similarity between a new sentence and another sentence already translated in thedatabase, it will suggest it to the translator with a percentage-rating of closeness.Benefits of TMSeveral compelling benefits can accrue to the organization with all (or even most) of its translation memory in a database.Analysis and cost estimates are more accurateAs observed somewhat ironically on page 2, the key metric in the cost of a localization project is Words. Before TM, it sufficed to estimate the wordcount of the entire project and multiply it by a price per word, but why pay tore-translate text that has already been translated, or which appears identically in many different places? With TM, it is possible to determine the wordcount of phrases and sentences that have already been translated, and thereby arrive at a more accurate cost estimate.Figure 5shows the analysis on a batch of ten new files submitted to a vendor for translation.• 38,891 words are in 4,130 repeated segments 1, or segments whichexactly match another segment in the ten files. The translator can translate the first occurrence of these segments and the TM software will propagate themthroughout the project. In addition, 4,674 words are ―placeable‖ (numbers, tags, symbols) and do not require translation.• In descending buckets from 100% down to 50%, there are a few hundred words in segments which have been translated before. These bucketsrepresent descending degrees of fuzzy match.• The software finds no match for segments containing 104,661 wo rds, so these must be translated from scratch. The 4,951 placeable words donot require translation.Many vendors offer discounts based on this analysis: The higher the percentage match bucket, the greater the discount on words in that bucket.• A 100% match means, of course, that no translation work is required, but the words in the 100% segments must still be ―touched‖ (engineering,desktop publishing, translation memory work, final review, QA), so fewvendors discount them entirely.• A 95-99% match often means that punctuation or the spelling of a single word has changed, or a word has been added or deleted, so a translator must do some light work on the segment and the vendor will discountthe segment slightly less.• Below 75% matches, however, discounts are less common because, by the time the translator has found and dealt with the differences between the old and the new text, s/he may as well translate it from scratch. The vendor can pre-translate new versionsBefore the new file has made it as far as the translator, the TM software will have pre-translated as many 100% matches as possible. For any segment that already has a match in the TM database, the software will retrieve and place the corresponding translation. This greatly reduces the translator’s work and shortens the time to deliver the completed product.In Figure 6, the Chinese segments marked with gold (lines 164-171 in the right-hand column) are 100% matches for the corresponding English segments on the left, which have been retrieved from TM and dropped into the translator’s work file in advance.Figure 6Translations enjoy leverage from one version to the nextSimilarly, the software identifies fuzzy matches and places them. The translator modifies the existing translation in light of how the source text has changed and adds the new segment to the TM database. This helps the translator spot English segments that have changed since the last time the product was translated.Figure 7In Figure 7, segment 174 reads ―You will be charged for this application.‖ The software found the closest fuzzy match (84%) and pre-translated it, tagging it green to call the translator’s attention to it. The TM Difference View window shows the very important change made to the English sentence since the last round of translation: The sentence now reads ―You will not be charged for this application.‖ The software provides enough of the original translation so that the translator does not need to start from scratch.The leverage from one version of the localized product to the next is a tremendous advantage of TM software. While it does not lower the overall wordcount of a product, it eliminates needless work for the translators and shortens time to market for the localized versions.All terms are available for lookup and concordance searchThe combination of the TM database and fuzzy matching also allows for concordance searches on specific text for similar, but not necessarily identical, occurrences.Figure 8In Figure 8 the translator has looked up ―Please try again later‖ in the TM database to see in how many different ways similar text has been translated in the past. This functionality goes deeper than that of a translation glossary because it broadens the subject of the lookup from key terms to common phrases, and the domain of lookup from the few hundred glossary terms to the entire TM database.Clients preserve history from one vendor to the nextFinally, the TM database represents a valuable asset if/when the time comes to change localization vendors. With TM, if a vendor goes out of business or is unable to scale to meet a client’s localization needs, the client can forward the TM database to the new vendor, who can then exploit the translation history with less delay.There may be technical limitations (different TM database formats) and legal issues (ownership of the TM database itself), but the larger the translation history, the smaller these issues look by comparison.A few more notes on TM• The real value in TM lies in its continued use over time. A sustainedinternational effort to deliver future versions of localized products willbenefit handsomely from TM. The organization interested in a one-time, quick-and-dirty translation will enjoy far fewer benefits.• Although TM saves a lot of work, it als o involves a lot of work for vendors.Some vendors bill for it directly and conspicuously, while others bill for it under general engineering costs. On balance, though, its benefitsoutweigh its costs, and there is almost no point in trying to save moneyby instructing the vendor not to use TM tools.• The client must bear in mind that translation memory tools are notmachine translation tools. As described above, MT tries to calculatetranslations of new text using rules and existing translations, whereasTM accumulates segment-by-segment history and assists humantranslators. While hybrid TM-MT solutions are becoming more popular,the industry is still a long way from reducing translation to a pure matter of software and hardware.• The client must b ear in mind the concept of the segment. TM looks forand matches text in entire segments because this is the lowest level ofgranularity which the software can use. For this same reason,wordcounts are a function of words in segments, as described in Figure5.Clients who do not understand the concept of the segment ask a verycommon question: ―My company and product name, Crunchware,appear 900 times in the product, and I want them preserved as―Crunchware‖ in all languages. Do I have to pay for that?‖ The answer is ―Sometimes.‖ If ―Crunchware‖ occurs as a segment all by itself in the TM database – and it probably does – then at all occurrences of thatsame segment there is no work for the translator to do, and the clientwill likely receive a steep discount for that 100% matched segment inevery place that ―Crunchware‖ appears alone. However, if―Crunchware‖ appears in a completely new sentence with 44 otherwords – like this sentence, for instance – then the software will report a 45-word segment with no match in TM, and the vendor will likely charge for 45 words at the full rate per word.SummaryA properly internationalized product is a delight to localize, because it involves no wasted effort in handoff, cost estimation, scheduling, translation, rebuilding, testing, release or support. It meets the needs of users in other regions with no changes to its core functionality, no patches, no bug lists, and no excuses. Enlightened organizations manage their own internationalization (I18n), which keeps it out of the black box.The black box looks black mostly because localization (L10n) takes place outside of the client’s organization and involves tools and skills rarely found inside the organization. The black box is not so much product development as product transformation, in which a familiar, English-original product becomes a French/Korean/Russian/Hebrew/... copy. Translation memory (TM) tools are at the heart of the transformation, keeping costs as low as possible and accelerating the work of translators.Although the L10n process seems trivial to many technology clients, there really is a great deal more to it than simply translating words from one language to another. Honest.。
本地化翻译(Localization)：DTP(Desktop Publishing桌面排版)工具简单介绍
专注翻译 恪守精准
图形处理工具（续）
Coreldraw (.cdr)：是目前使用普遍的矢量图形绘 制及图像处理软件之一，该软件集图形绘制、平面 设计、网页制作、图像处理功能于一体。 FreeHand (.fh8, .fh9, .fh10, .fh11)：是一款非常强 大平面向量绘图软件，同時支持 Mac OS 及 MS Windows 平台，最早由 Altsys 公司开发，是 Adobe Illustrator 的强劲竞争对手，但几经周折， Adobe 公司通过收购 Macromedia 将其纳入囊中 ，并几乎停止了此软件的更新。
专注翻译 恪守精准
音频处理工具பைடு நூலகம்
音频处理工具： Adobe Audition 等
Adobe Audition (.wav, .mp3)：是一个专业音频编辑 和多轨道混音环境（它的前身是大名鼎鼎的 CoolEdit ）。 Adobe Audition 专为在广播和后期制作方面工作 的音频和视频专业人员设计，可提供先进的音频混合、 编辑、控制和效果处理功能。最多混合 128 个声道， 可编辑单个音频文件，创建回路并可使用 45 种以上的 数字信号处理效果 — Audition 是一个完善的多声道录 音室，可提供灵活的工作流程并且使用简便。无论要录 制音乐、无线电广播，还是为录像配音，Audition 中 的恰到好处的工具均可为您提供充足动力，以创造可能 的最高质量的丰富、细微音效。
专注翻译 恪守精准
文字处理工具（续）
InDesign (indd, inx)：其定位是面向创意设计领 域的专业设计、排版与跨媒体编辑工具。具有强大 的页面排版功能、灵活方便的表格功能、丰富的图 形图像处理能力，可与 Adobe 系列产品中的其他 产品紧密集成。其特点是功能强大、应用广泛、兼 容性好（ InDesign 软件可直接打开 PageMaker 6.0-7.0 版本的文件及 QuarkXPress 3.3-4.x 版本 的文件，对于因缺少相应语言版本而无法进行处理 的 PageMaker 或 QuarkXPress 项目，如果客户 接受，便可转用 InDesign 来完成）。
翻译行业的本地化翻译总结
创新发展
鼓励本地化翻译技术的创新和应用，推动行业的数字化转型和升级 ，实现可持续发展。
05
案例分享
案例一：游戏行业的本地化翻译
总结词
游戏行业本地化翻译需注重文化差异和玩家体验。
详细描述
游戏行业本地化翻译不仅仅是简单的语言转换，还需要考虑到不同地区玩家的文化背景和游戏习惯。在翻译过程 中，需要注重游戏内容的准确传达，避免因文化差异引起的误解或不良体验。同时，还需关注游戏界面的交互性 和易用性，确保玩家在游戏过程中能够顺利理解和操作。
THANKS
感谢观看
专业术语准确
对于涉及特定领域的文本，确保使用准确、专业的术语，并进行适 当的注释或解释。
技术问题的解决
翻译工具使用
熟练掌握各种翻译工具和软件，以提高翻译效率和准确性。
术语管理
建立和维护翻译术语库，以确保译文中术语的一致性和准确性。
格式与排版
根据目标市场的排版要求，对译文进行适当的格式调整和排版， 以确保译文的易读性和美观度。
确保译文准确传达原文意 思，不出现错译或漏译。
语言流畅
译文需符合目标语言的表 达习惯，流畅自然。
后期处理与发布
排版调整
根据目标市场的排版规范 ，调整译文格式。
适应性测试
在小范围内测试译文，确 保在不同设备和浏览器上 显示正常。
多语种发布
将译文发布到多个语种平 台，扩大覆盖范围。
质量保证与反馈
质量标准制定
制定详细的质量标准，确保每个 环节达到预期效果。
定期审查
对完成的翻译进行定期审查，确 保质量稳定。
用户反馈收集
本土化翻译
本土化翻译一、本土化翻译的定义与重要性本土化翻译，指的是在翻译过程中，不仅要准确传达原文的信息和意义，还要使译文在目标语言文化中得以自然、顺畅地传播和接受。
它强调在翻译过程中，充分考虑目标语言读者的文化背景、语言习惯和接受能力，使译文符合当地的风俗习惯和审美标准。
本土化翻译的重要性不言而喻，它能够有效提升译文在目标市场的接受度，扩大产品或服务在国际市场的竞争力。
二、本土化翻译的策略与方法1.文化适应：在翻译过程中，对原文进行适当的调整和修改，使之符合目标语言的文化特点。
例如，一些具有浓厚文化背景的词汇、成语和谚语，需要在翻译时加以解释或寻找与之对应的目标语言表达。
2.语言习惯调整：在保证译文准确性的基础上，尽量使译文符合目标语言的语法结构、词汇搭配和表达方式。
例如，英语和汉语在表达方式上存在很大差异，翻译时要注意调整句子结构和语序，使译文更自然。
3.信息增删：在保证原文信息传递的前提下，根据目标语言的表达习惯，对原文进行适当的增删。
例如，一些在我国特有的政策、制度或术语，在翻译时可以进行简化或概括，以便目标语言读者理解。
三、我国本土化翻译的现状与发展随着我国国际地位的不断提升，国内外交流日益频繁，本土化翻译在我国也得到了广泛关注和发展。
目前，我国本土化翻译已涉及到多个领域，如文学、影视、广告、企业文档等。
然而，与西方发达国家相比，我国本土化翻译在理论研究、技术应用和实践经验方面仍有较大差距。
因此，本土化翻译在我国仍有很大的发展空间。
四、本土化翻译案例分析以我国电影《哪吒之魔童降世》为例，其在国外的推广和传播过程中，进行了充分的本土化翻译。
不仅在剧情、人物设定等方面进行了文化适应，还将中国特色的元素如神话传说、传统服饰等进行了恰当的呈现。
此外，该片在台词和歌词翻译上，也尽量保持了原文的风趣幽默，使国外观众能够更好地感受到中国电影的魅力。
五、提升本土化翻译能力的建议1.加强本土化翻译理论研究：深入探讨本土化翻译的原理、方法和技巧，为实践提供理论支持。
本地化行业(Localization)综合类专业术语
综合 综合
产品本地化实施过程中相互联系、相互作 用的一系列过程。
产品本地化流程中的输入输出的对象。
综合
提供本地化服务的类别。
Common Sense Advisory 评选“世界百强语言供应商”︱专注翻译 恪守精准 2
综合
不需要重新设计或修改代码、将程序的用 户界面翻译成任何目标语言的能力。本地 化能力是表征软件产品实现本地化的难易 程度的指标。
Technology of localization
综合
产品本地化过程中应用的各项技术。
多字节ቤተ መጻሕፍቲ ባይዱ符集
Multi-byte Character Set
质量保证 文件格式 用户界面 服务角色
软件中与用户交互的全部元素的集合，包 括对话框、菜单和屏幕提示信息等。
产品本地化实施过程中承担不同任务的各 种角色。
Common Sense Advisory 评选“世界百强语言供应商”︱专注翻译 恪守精准 1
服务流程 服务要素 服务种类
Process of service Element of service Types of service
中文术语 本地化 国际化 全球化 本地化能力 本地化技术
本地化行业(Localization)综合类专业术语
英文术语 Localization (L10N) Internationalization (I18N) Globalization (G11N)
类别 综合 综合 综合
定义
将一个产品按特定国家/地区或语言市场 的需要进行加工，使之满足特定市场上的 用户对语言和文化的特殊要求的软件生产 活动。
在程序设计和文档开发过程中，使功能和 代码设计能够处理多种语言和文化传统， 从而在创建不同语言版本时，不需要重新 设计源程序代码的软件工程方法。
本地化翻译的语言文字表达规范
本地化翻译的语言文字表达规范一、本地化译文的基本准则∙凝练平实，言简意赅；∙信息全面，含义准确；∙语气流畅，逻辑通顺；∙使用书面用语，符合汉语语法习惯；∙杜绝错字、别字、多字、少字、标点符号误用和英文拼写错误；∙译文的用词及语气须避免有对性别、年龄、种族、职业、宗教信仰、政治信仰、政党、国籍、地域、贫富以及身体机能障碍者的歧视。
作者：不详原始出处：本地化世界网搜索更多相关主题的帖子: 语言文字本地化翻译规范TOP软件测试从业人员3G普及现状有奖调查青铜元老∙个人空间2#大中小发表于2007-11-3014:05 只看该作者本地化的项目绝大多数属于科技英语的本地化项目，在科技英语项目的本地化翻译时要掌握以下语言规律和特点：从文体上看，大多是论述性、指南性的，多用陈述句、祈使句，平铺直叙，少有感情色彩。
∙发短消息∙加为好友∙当前离线句子结构简练严谨，常采用省略手法，用短语来代替从句。
词汇力求短小精悍，常用复合词，技术性越强，复合词越多。
在表现手法上力求客观性，避免主观性和个人色彩，被动语态使用较多，以使句子紧凑，主语信息丰富，避免重复。
文章结构层次分明，连接词的使用十分频繁和重要。
用词比较正规。
注意：手册的的语言风格与联机帮助或界面相比要略显活泼一些，经常会出现一些疑问句、反问句、感叹句、俚语等；在翻译时要将这些地方译得文雅而不口语化，既要传达出原文要表达的感情，而表达方式又符合汉语的习惯。
TOP每周一问：如何构建大量的测试数据？青铜元老∙个人空间∙发短消息∙加为好友3#大中小发表于2007-11-30 14:05 只看该作者名称和地址的翻译规则：∙人名、地址、公司及客户名称o原文中虚拟的人名、地址、公司名称及客户名称若译成中文，应避免与名人或真实的公司名称有雷同的情况，亦不得谐音。
o地址名在需要的情况下也请使用中文。
如有疑问，请IQA或该产品组查询确认。
∙产品名称o原则上正式上市的中文版Microsoft产品，其产品名称均维持原文格式，不加以翻译。
本土化翻译
本土化翻译一、本土化翻译的定义与重要性本土化翻译，顾名思义，是指将某个产品、服务或理念从原产地传播到其他国家时，对其进行符合当地语言、文化和习俗的翻译。
这种翻译方式不仅仅局限于词汇的转换，还包括对原文的文化背景、语境和表达方式进行调整，使其在目标市场中更具吸引力和可接受性。
本土化翻译的重要性不言而喻，它是推动跨国企业、国际合作和跨文化交流的重要工具。
二、本土化翻译的策略与方法1.语言适应：在本土化翻译过程中，译者应充分了解目标语言的语法、修辞和表达习惯，以确保译文的准确性和地道性。
2.文化适应：翻译者需要深入研究目标文化，对比原文化和目标文化之间的差异，对原文进行适当的文化调整，使其符合目标市场的审美和价值观。
3.语境适应：在翻译过程中，译者要充分考虑原文的语境，包括场景、目的和受众等，以确保译文在目标市场具有针对性和实用性。
4.品牌传播策略：在本土化翻译中，译者还需注重品牌传播策略，保留原文的品牌形象和核心价值，同时使其在目标市场具有吸引力。
三、我国本土化翻译的现状与发展趋势随着我国经济的快速发展和国际地位的日益提高，本土化翻译在我国的应用越来越广泛。
目前，我国本土化翻译主要集中在企业国际化、文化传播和对外援助等领域。
然而，相较于国际先进水平，我国本土化翻译仍存在一定差距，表现在译文质量、翻译技术和人才培养等方面。
近年来，我国本土化翻译行业逐渐呈现出以下发展趋势：1.技术驱动：人工智能、大数据等先进技术在本土化翻译领域的应用，提高了翻译质量和效率。
2.专业化分工：随着市场需求的增长，本土化翻译行业逐渐形成了以专业翻译公司、自由译者和语言服务提供商为主体的多元化竞争格局。
3.国际化合作：我国本土化翻译企业逐步拓展国际市场，与世界各地的翻译机构开展合作，促进全球翻译资源的共享与整合。
四、提升本土化翻译质量的途径1.加强人才培养：提高本土化翻译质量，首先要加强对翻译人才的培养，提升译者的语言能力、文化素养和专业素养。
- 1、下载文档前请自行甄别文档内容的完整性，平台不提供额外的编辑、内容补充、找答案等附加服务。
- 2、"仅部分预览"的文档,不可在线预览部分如存在完整性等问题,可反馈申请退款(可完整预览的文档不适用该条件!)。
- 3、如文档侵犯您的权益，请联系客服反馈,我们会尽快为您处理(人工客服工作时间：9:00-18:30)。
专注翻译 恪守精准
翻译症 – 解析 2
英文和译文
• The aim of the EIT Remote audit and reporting is to support the Owner with efficiency analysis on a long term basis to increase and maintain the productivity.
• 译文1： XXX 系列交流变频器功能强大，具有易于
使用、灵活且适用于各种工业应用的特点。 • 译文2： XXX 系列交流变频器坚固耐用，其易用性、 灵活性和极佳的性能适用于各种工业应用。 • 译文3： XXX 系列交流变频器坚固耐用，兼具易用、 灵活和高性能特点，适合各种工业应用。
存在的问题 理解和表达中都存在问题： 1. 译文过长，主要由于修饰 成分过多，状语和定语并 列显得过多，造成句子冗 长。 2. to increase and maintain the productivity 本为目的 状语，译文中翻成了定语
7
• EIT 远程审计和报告的目标 是要通过长期基于提升和保 持生产力的效率分析来为所 有者提供支持。
2
专注翻译 恪守精准
翻译的标准
• 信、达、雅 – 信 - 忠实 – 达 - 通顺 – 雅 - 风采 • 直译和意译 • 美学取向的翻译原则
ห้องสมุดไป่ตู้
3
专注翻译 恪守精准
翻译症 – 语言表达
• The XXX Series is a robust family of AC drives that provide ease of use, flexibility, and performance for a variety of industrial applications.
4
专注翻译 恪守精准
翻译症 – 通顺性
• For functions with a similar syntax, a selected function is used as a template in the example. • 译文1：对于语法类似的函数，示例中将精选 出一个函数作为模板。 • 译文2：对于语法类似的函数，示例中选择了 其中一个函数进行说明。 • This example must be adapted as well. • 译文1：还必须对示例进行调整。 • 译文2：用户应根据实际需要对示例进行调整。
如何做好翻译？
Empower Communications
专注翻译 恪守精准
专注翻译 恪守精准
译者应具备哪些素质？
• 扎实的语言基本功。译者应具有较强的驾驭译出语和译入语的 能力，能正确理解原文，熟练运用本族语。 • 具有丰富的文化知识，熟悉以英语为母语的国家的诸如历史、 宗教、政治、地理、军事、外交、经济、文艺、科学、风土人 情、民俗习惯等方面的社会文化。这有助于正确理解原文，完 美地表达原作所要传达的思想内容。 • 熟悉翻译理论和常用技巧，善于灵活运用各种翻译技巧。 • 熟悉各种工具书。译者的记忆力和知识面终归是有限的，因此 遇到问题就应该知道怎样去查找正确的答案，从哪本书中去寻 找。这就要求他应该知道怎样使用各种工具书。常见的工具书 有各种单语词典和双语词典、百科全书、史书等。译者应该熟 悉这些工具书各自的特点和用法，在翻译时如遇到难题就可求 助它们。
5
专注翻译 恪守精准
翻译症 – 专业知识
• Easily assessable heat sink and internal fans. • 译文1：易于装配的散热片和内部风扇 • 译文2：散热器和内置风扇易于检查
9
专注翻译 恪守精准
北京 • 沈阳 • 上海 • 成都 • 香港 • 台湾 • 日本 • 韩国 • 新加坡 • 美国 • 德国
解决方法
• 转换句子成分 • 定语过长，整个句子结构不 平衡，没有进行翻译后的复 查，重新认真阅读一遍的过 程没有，或者没有带着重新 审视的角度复查，只是看看 是否有语法错误和低级错误。 作为合格的翻译就应该将复 查作为提升质量的关键一步。
• EIT 远程审计和报告的目标 是要通过长期基于提升和保 持生产力的效率分析来为所 有者提供支持。
8
专注翻译 恪守精准
翻译症 – 解析 3
• The aim of the EIT Remote audit and reporting is to support the Owner with efficiency analysis on a long term basis to increase and maintain the productivity. • 译文1：EIT 远程审计和报告的目标是要通过 长期基于提升和保持生产力的效率分析来为所 有者提供支持。 • 译文2：EIT 远程审计和报告的目的是通过效 率分析长期为“所有者”提供支持，从而提高 并保持生产力。
6
专注翻译 恪守精准
翻译症 – 解析 1
英文和译文
• The aim of the EIT Remote audit and reporting is to support the Owner with efficiency analysis on a long term basis to increase and maintain the productivity.