Skip to main content

Glossary generator

  • Updated
Content is translated from English using DeepL Pro.

Is there a way to generate glossaries automatically, based on existing translation work?

Yes, the glossary generator allows users to quickly and easily generate their own glossaries based on previous translations. With just a few clicks, upload previously translated files and the glossary generator extracts potential translation rules to collect in a new glossary you can use in DeepL. 

In this article:

 

What is the glossary generator?

The glossary generator is a feature that helps simplify and speed up the process of creating glossaries for your translations.

The glossary feature in Translator allows you to easily set and apply translation rules for words and short phrases, in order to ensure consistency in your translations across an entire team.

To create and enrich your glossary, you can add new entries manually or upload existing glossaries in CSV format. With the glossary generator, you can also use previously translated files to automatically generate entries for your DeepL glossary, eliminating manual work.

Important: Using files translated only via machine translation might lead to unsatisfactory results when generating a glossary. We recommend using files that were translated or reviewed by a human editor.

The glossary generator is currently only available through our website. It is not yet included in our desktop apps, mobile apps, browser extensions, or our API.

To access and use the glossary generator, you'll need a DeepL Pro Advanced or Ultimate plan (or a bundle that includes one of these 2 plans). It is currently unavailable for users of our free service and users with a DeepL Pro Starter, DeepL API Free/Pro, or DeepL Write Pro plan. You can find more information about our plans here.

 

Supported language pairs

  • English (US & UK) ↔ German
  • English (US & UK) ↔ Spanish
  • English (US & UK) ↔ Japanese
  • English (US & UK) ↔ French
  • English (US & UK) ↔ Italian
  • English (US & UK) ↔ Russian

More languages will follow soon.

 

How to generate glossary entries

You can create new glossaries with the glossary generator from the following file types:

  • .docx
  • .pdf
  • .tmx (Translation Memory eXchange)*
  • .tsv
  • .csv

More file types will follow soon.

Before you start, log in with your DeepL account.

To access the glossary generator:

  1. Go to our web translator and click Glossary
  2. If you don't have glossaries yet: click on Upload and then on Other file types.
    If you already have glossaries: click Manage glossaries and then Generate in the glossary window.

Before you begin, select the file type from which you'd like to extract potential glossary entries by clicking the corresponding tab.

To generate a glossary:

  1. Click Browse to select a file or drag and drop it to the page
    1. For .docx and .pdf: Select one file containing the original text and another file with its translation 
    2. For .tmx, .tsv. and .csv: Select a file containing both the original text, and its translation
  2. Select a source and target language under Language pair
  3. Click Submit

After the upload, the glossary generator will extract candidates for glossary entries (or "translation rules") from your files. This might take a while, depending on the size of the files.

As soon as the glossary entry candidates are ready, you'll receive an email. If you don't receive an email, please check your spam folder.

The email will include a link to a page containing the newly generated glossary entries. On this page, you can:

  • Edit entries
  • Delete entries
  • View example sentences with the suggested terms
  • View the Relevance score of an entry (the higher the score, the more often a term appeared in the uploaded files)
  • Save your progress
  • Name your new glossary before generating it

After you've finished editing your list of glossary entries, you can either download the glossary as a CSV file, or directly click Generate glossary to finalize and instantly add the new glossary to DeepL. 

 

* .tmx is a file format that is used in the translation industry to store and exchange translation memories. Translation memories contain previously translated text pairs that can be reused.

 

Glossary generator limitations

Note that there are glossary limits depending on your plan type. More information about glossary limits can be found here.

You can only upload one .tmx/.tsv/.csv file or one pair of .docx/.pdf files (original text and its translation) at a time. Each file shouldn't exceed 500 MB in size. You can run up to 3 glossary creation processes in parallel.

Related articles