Introduction to translation with Lingo24
Thank you for choosing to translate with Lingo24. We’ve made a significant investment in technology to streamline the process of translation and put the power in the customers hands. Below is some quick start information to help you get started.
This section will help you understand the translation process and select the language combinations and services you need for each order, ensuring your orders can be translated, reviewed and delivered on time.
When you upload files for translation (we call these source files), we extract the text content to be translated, while protecting content which should not be translated (e.g. styles and formatting, in-line code, numbers and placeholders).
So if you upload a Microsoft Word (.docx) file we will extract all of the visible editable text from the file including the headings, paragraphs and even the text from cells within tables. Content from assets such as images is not translated by default.
A key feature of the Lingo24 translation is our ability to control what should be translated and what should not, and automate this for those who order regularly, so it doesn’t need to be specified each time. See ‘What if I don’t want all of the content in my file translated?’ and ‘What if I want images translated / recreated?’.
The extracted text is then separated into smaller chunks called segments – often these are individual sentences from the text. Segments are then checked for repetitions and then checked against your translation memory to see whether these have previously been translated.
This defines the scope of the order by working out what needs to be translated afresh and what can be reused (we call this being “leveraged” from previous requests.
Your translation memory (TM) is a database of segments that have previously been translated.
Where new source segments have previously been translated these translations are retrieved from the translation memory and populated in the translation target segments (this is a process known as leveraging).
As well as whole segments which have previously been translated, the translation memory can also pick up where there’s a close match to a segment which has previously been translated. These are known as ‘fuzzy matches’ and are split into categories based on the percentage of text which matches.
Populating the translation target segments helps to maintain consistency and brand voice, reduces the time required to translate content and reduces the cost of your translation.
When we have leveraged the translation memory we count the number of leveraged words and the number of new words and we use these counts to calculate the cost of your translation.
The translation and review process all happens in our Computer Assisted Translation (CAT) tool software.
This software helps linguists to translate your content and includes a number of quality control features for quality and consistency. As well as the content of your files the linguists also have access to your whole translation memory and terminology (an approved set of terms which should or should not be used) to help make translations consistent and on-brand.
Linguists do not translate directly into your target file, however they do have access to your source file for reference. Once approved by the reviewer all of the translated segments will be stored in your translation memory so that they can be used for future projects.
When the translation is complete we extract the text from our Computer Assisted Translation (CAT) tool and merge it into a new file with the formatting from your original source file.
So if you uploaded a Microsoft Word (.docx) file in English (UK) for translation into French (France) and German (Germany) you will receive two new Microsoft Word (.docx) files back, one in French (France) and one in German (Germany).
Depending on the formatting in your document and the relative length of the target language you may find that some text does not fully display in text boxes, tables, pages etc, you can either correct this yourself manually or request Lingo24 to perform these adjustments for you using our Desktop Publishing service, if you would like to add this to your order.
Documents translated into different languages can have text expansion (where a higher number of words or characters are used) or text contraction (where a lower number of words or characters are used). This varies per language and can have a significant impact on your design.
For documents where the layout is important (e.g. Microsoft PowerPoint presentations or documents that have been created using Desktop Publishing software such as Adobe InDesign), the target files can be reviewed and the layouts updated to ensure that all content is correctly displayed on pages and within text boxes and images.
If you would like Adobe InDesign (.idml) or Microsoft PowerPoint (.pptx) files returned fully formatted you can use our typesetting services. These are our Full DTP service (for Adobe InDesign files) or our Format as Source Service (for Microsoft PowerPoint files).
When creating a translation order it is important to select the correct service and language combinations to ensure that the correct translation memory is used to provide you with consistency and accurate pricing.
- Selection of the translation memory and pricing
- Identifying which of our native translators are able to work on the project
- Controlling the assets used – i.e. spellchecker, glossary, MT engine – as this can very greatly across variants.
Selecting incorrect services and languages may lead to delays in your translation delivery and cause quality issues.
For example, each of the orders below would use separate translation memories:
- English (UK) to Spanish (Spain) – Professional Translation service
- English (US) to Spanish (Spain) – Professional Translation service
- English (US) to Spanish (Lat Am) – Professional Translation service
- English (US) to Spanish (Lat Am) – On-brand Translation service
If in doubt, your Account Manager can help you make the choice most suitable for your projects.
You may have content in your source files that you do not want translated, for example:
- In a Microsoft Word (.docx) file you may not wish to translate the headers and footer content.
- In a Microsoft Excel (.xlsx) file you may not wish to translate the content in all columns or sheets.
- In a Microsoft Powerpoint (.pptx) file you may not wish to translate the slide notes.
If you do not want all of the text content in your files translated you have two options:
Option 1: You can edit your file and delete or hide any content that you do not want translated before uploading your file for translation.
- For example, you can hide columns or tabs using the hide feature in Microsoft Excel.
Option 2: Lingo24’s File Engineering team can filter out content that you do not want translated before the file is counted.
- For example, when you upload your Microsoft Word (.docx) file you can ask us not to translate any content styled using Heading 1 and Heading 2.
- Just make sure that the content that you do not want translated is always styled in a consistent manner and follow the instructions in “Step 2.1. File engineering for non-translatable content”.
You may have words or phrases (such as client, brand or product names) that need to be in the final translated document and also left in English. In these cases, you should not delete, hide or highlight these.
Where you have words or phrases that must be left in English add this information in the Project Instructions on when you place your order and we will share this with your Translation team. We can even add these to your Terminology in our Computer Assisted Translation (CAT) tool to ensure that these always remain in English.
Text contained within images is not extracted and translated as a standard. If you would like text extracted from images and translated and if you would like your images recreated in your target languages you can add this information by requesting Desktop Publishing.