Language Scientific® utilizes Translation Memory (TM) to help you maintain consistency while reducing overall costs and turnaround times. As a US-based technical and scientific translation and localization provider, we serve over 1,500 global corporations in more than 215 languages. We offer a full suite of language services, helping you deliver precise global communication. Translation Memory is part of the suite of Computer-Assisted Translation (CAT) tools we use to enhance your localization project.

How Does Translation Memory Work?

Translation Memory (also called TM or translation database) is a segment-by-segment database storing every translation to be recycled and reused to maintain quality and consistency while enabling multiple Knowledge Experts to work concurrently, reducing your overall translation and localization costs and speeding delivery.

We use some of the best technology in the industry to enhance quality, reduce the turnaround time for translation, and lower costs. Our Translation Memory tools monitor the progress of translation in real time and memorize each translated passage. Then, when a paragraph or sentence is encountered that has been translated before—possibly years earlier by a different translator working on the same project—the TM notifies the translator(s), allowing them to insert or modify the pre-translated text.

Customized Segmentation For Translation Memory

At Language Scientific, our segments can be customized to meet the needs of your project(s). The Translation Memory stores language pair phrase segments from previous translations. Segments are phrases, lines, sections of text or sometimes even complete sentences. We can customize your segments to be set for certain punctuation (for instance to begin at a period or semicolon or at a hard return) or any other way that will optimize your project.

Translation Memory To Maintain Consistency

Our Translation Memory tools help you achieve consistency and uniformity, because the same phrases are translated identically and the terminology is used consistently even if different translators work on different portions of the document.

 Fuzzy Matching To Save Translation Costs And Turnaround Times

Language Scientific applies fuzzy matching to your existing Translation Memory to reduce translation costs and turnaround times and to deliver you optimal translations with minimal expense. We use fuzzy matching to analyze your quotes to best determine cost and time for performing the project. Fuzzy matching is most useful when you update your translated or localized content on a regular basis.

We perform a search where your new content is compared against the TM’s segments to find matched content. We evaluate each segment in the Translation Memory and compare it with the new content to generate a report detailing the text additions, repetitions, deletions and modifications of the existing text.

Once the report is received, new text is designated for translation, deleted text is removed, 100% perfect matches are eliminated so that time and resources are not spent re-translating the same content. Fuzzy matches are returned on a scale from 50% up to 99%. The higher the percentage of match, the more useful it is to your translation team to speed up the translation.

When a fuzzy match is found for a particular segment, the TM suggests a translation that the translator can accept, reject or modify to meet the needs of the target language. This increases consistency across your translation and can significantly reduce your translation or localization project’s costs and turnaround time.