Technical Topics

From file formats to automation options

Which file formats can be localized?

We work with all the common file formats. However, for analysis and the majority of localization and DTP work, it is essential that we receive live source files (i.e. editable files in their original file format) and not only a PDF of the document/content in question.

How are graphic texts, websites, and software files localized?

With the help of advanced localization software, we extract all of the text from the live source files. Text segments are then translated by qualified specialized translators using localization software. At the end of the process, the translated text segments are reimported into the coded files.

What is CAT?

CAT (Computer-Aided Translation) refers to the use of translation memories, terminology databases, or other machine translation software as tools to aid the translation process.

What is a translation memory?

A translation memory (TM) is a database that contains existing translations for reuse/leveraging. The use of translation memory databases is part of the IF-Localization standard workflow. Leveraging translations from a TM reduces translation cost and enables shorter turnaround times, since sentences and phrases that have already been translated as part of a previous project — now stored in the TM — can be reused. The use of a TM also ensures that text that occurs repeatedly — within a project or over numerous projects — is translated consistently.

What is alignment?

Alignment is the process of preparing existing translations, not yet contained in a translation memory, for importing into such a database. For every language pair, the source sentences (also called segments) are coupled with the corresponding target sentences. Alignment software is used to pair the segments automatically, followed by a thorough check by a translator who manually corrects and adjusts the alignment as needed.

What is terminology extraction?

During terminology extraction, terminology candidates contained in a body of text are identified, extracted, and compiled in a list for further processing. With the help of terminology extraction software, a list of potential term candidates is created. A linguist then reviews the extracted terminology candidate list and edits the results manually.
Term extraction can be performed on the source language or simultaneously on both the source and target languages. For automated multilingual extraction, aligned, bilingual, parallel texts are required for each language pair.
Terminology extraction output can be used to create a new terminology database or be used to expand existing terminology databases.

What is controlled language?

Controlled language means restricting (i.e. controlling) language use by means of specific grammar rules, predefined terminology, and stylistic guidelines. Stylistic guidelines, for example, could stipulate a maximum text length or the exclusive use of active sentence structure.
Rule-based software programs for controlled language are being used more and more frequently, especially in technical communications. The goal is the creation of consistent, non-ambiguous text that is easier to understand and thus more user-friendly, and text that is easier to translate, thereby reducing the associated translation costs. Controlled language is also an interesting factor for the future of machine translation.

What is machine translation? How good is the quality?

Machine translation, based on linguistic rules and statistical analyses, uses software alone for the translation of texts.
In order to achieve good results using machine translation, extensive preparatory ground work is required for every subject area and every language. The resulting translation must still be reworked by a human translator to ensure the needed quality. For these reasons, machine translation is not, to date, considered a cost-efficient solution in most cases.

Do you have any questions? Would you like to receive a quote? Are you interested in more detailed information? Please contact us!


Luka Komljenović

Would you like to know more?
We look forward to your inquiry.

> +49 (0) 89 / 610 49-301
> Luka.Komljenovic(at)