How to Create a Quality Translation Memory

如何建立高质量的翻译记忆库?

2020-08-04 00:10 Memsource

本文共673个字,阅读需7分钟

阅读模式 切换至中文

How to Create a Quality Translation Memory Translation memory (TM) is essential for producing consistent translations and can dramatically reduce translation spend. However, if your TM is not set up correctly from the get-go, and well maintained, you run the risk of producing inconsistent, poor quality translations that could harm your brand image, deliver misinformation, and accumulate unnecessary costs fixing errors. We’ve put together some handy steps to follow to ensure you create quality translation memories that are easier to manage and maintain. 1. Choose your providers wisely If you have a group of trusted providers (linguists/vendors) who deliver high-quality output you can let them save their translations directly to your master TM. However, when you work with a provider for the first time or with someone whose output quality varies, it is better to consider using a secondary “working” translation memory where they can commit segments and use the master TM in read-only mode. The master TM can be used in Read and Write mode in the later workflow steps where review is performed. ​ It is easier to prevent content of questionable quality from being included in your TM from the start than it is removing it later. Proposed configuration of translation memories when working with a new vendor. 2. Add context information to your source files Context information allows your linguists to better understand the content they are translating, which improves the quality of the translation. Depending on what tool you’re using, there may be different options to provide context—you can attach assets as reference files to projects or add them on the segment level. For file formats with context key and notes properties, the information can be displayed on segment level in the CAT tool. Some editors are even able to display animations and graphics from attached external links. 3. Lock segments with high-quality matches Pretranslating content from translation memory and locking high-score matches (context matches) will prevent unwanted changes in your TM. Some systems also allow you to exclude locked segments from analysis and quotes that you share with your provider, reducing translation volume and costs. 4. Perform Quality Assurance and spell checks before confirming to TM It’s human nature to make mistakes. Misspellings, missing tags, incorrect punctuation are easily overlooked. Automated Quality Assurance (QA) checks can help with this. More advanced QA checks are also able to verify if correct terminology has been used—ensuring translation consistency. Some tools enable segment-level QA which won’t allow the provider to confirm segments and save them into the TM if Quality Assurance errors have been found. In case segment-level QA is not available (and the check is performed at the end of the localization process), it would be advisable to use the working TM approach mentioned above. 5. Perform Linguistic Quality Assurance (LQA) evaluation LQA evaluation can be used to measure and qualify the translations and errors produced. It’s a great way to evaluate translation quality and provide constructive feedback to the provider. 6. Update your TMs with any changes that happen outside of your translation management system If any linguistic edits take place in the native format or directly in a content management system, they won’t be saved in your translation memory and will be overwritten by future submissions of the same content unless the TM is updated. In such a scenario, it is important to update the TM manually. 7. Close the feedback loop Discuss the quality of delivered translations with your provider and allow them to see the changes made to their work. It is important to clarify your expectations and review detected issues to avoid encountering them again in the future. Conclusion Translation memory is a powerful tool that can boost your productivity, but this is only possible if you have high-quality translation memories. Otherwise, it will have the opposite effect and end up creating additional work for your team. Use these tips to craft a first-rate translation memory strategy and you will quickly reap the rewards.
如何建立高质量的翻译记忆库? 翻译记忆库是确保术语一致的关键,可以大大减少翻译成本。 然而,如果您的翻译记忆库从一开始就没有正确地设置,并没有得到很好的维护,那么您就有可能面临术语不一致、翻译质量差的风险,这可能会损害您的品牌形象,传递错误信息,并造成不必要的错误修复成本。 我们整理了一些简单的步骤,以确保您创建出更易于管理和维护的高质量翻译记忆库。 1.选择值得信赖的语言服务供应商 如果您有一群提供高质量译文且值得信任的语言服务供应商(语言学家/语言供应商),则可以将其翻译直接保存到您的主翻译记忆库中。 但是,当您初次与某位提供商或译文质量不稳定的人一起工作时,最好考虑使用辅助的“协同”翻译记忆库,他们可以提交语言字段并以只读模式使用主翻译记忆库。 在稍后执行检查的工作流程步骤中,可以在读取和写入模式下使用主翻译记忆库。相比最后再进行错误删除,从一开始就避免将质量有问题的内容保存到翻译记忆库中会省事很多。 与新的语言服务供应商合作时,明确翻译记忆库的要求。 2.在源文件中添加上下文信息 上下文信息可以帮助译员更好地理解他们正在翻译的内容,这便能提高翻译质量。 根据您使用的工具,可能会提供不同的选项来提供上下文。您可以将资产作为参考文件附加到项目中,或将其添加到细分级别。 对于具有上下文关键字和注释属性的文件格式,可以在CAT工具中以段级别显示信息。 一些编辑器甚至能够显示来自附加外部链接的动画和图形。 3.用高质量的匹配锁定片段 从翻译记忆库中预翻译内容并锁定高分数匹配项(上下文匹配项)将防止TM发生不必要的更改。 某些系统还允许您从与供应商共享的分析和报价中排除锁定的句段,从而减少翻译量和成本。 4.在确认翻译记忆库前执行质量保证和拼写检查操作 犯错是人之常情。 拼写错误,标签缺失,标点符号错误很容易被忽视。 自动质量检查(QA)可以帮助实现这一点。 更高级的质量检查也能够验证是否使用了正确的术语--确保翻译的一致性。 有些工具支持段级质检,如果发现质检错误,则不允许提供商执行确认该字段以及保存的操作。 如果区段级质检不可用(同时检查是在本地化过程结束时执行的),建议使用上面提到的工作TM方法。 5.进行语言质量保证(LQA)评估 LQA评估可用于衡量和确认所产生的翻译和错误。 这是评估翻译质量并向提供商提供建设性反馈的好方法。 6.通过更新系统之外的改变来更新翻译管理系统 如果任何语言编辑都是以原生格式或直接在内容管理系统中进行的,那么它们将不会被保存在您的翻译记忆库中,并且将被未来提交的相同内容所覆盖,除非更新了 翻译记忆库。在这种情况下,手动更新翻译记忆库非常重要。 7.关闭反馈回路 与您的提供商讨论已交付翻译的质量,并允许他们查看对其工作所做的更改。 重要的是,明确您的期望和检测过程中遇到的问题,以避免再次遇到此类问题。 结论 翻译记忆库是一个强大的工具,可以提升你的工作效率,但这只在你有高质量的翻译记忆库的基础上才能实现。 否则,就会产生相反的效果,最终给您的团队造成额外的工作。 使用这些技巧来制定一个一流的翻译记忆库管理策略,你定会大有收获。

以上中文文本为机器翻译,存在不同程度偏差和错误,请理解并参考英文原文阅读。

阅读原文