Getting started

Overview

An overview of the dashboard translator, document jobs, editors, language assets, and review tools.

DeepReference translates text and documents on its own EU-hosted model infrastructure. The workspace also includes segment and DOCX editors, Translation Memory, glossaries, terminology extraction, revision comparison, and QA checks.

For a short request, paste text into the dashboard and choose a target language. A document job adds segmentation, file rebuilding, status, and editor access. Approved segments can be stored for reuse in later work.

The four places you will work

• The dashboard translator: translate text or upload documents. This is the starting point for everything.
• The Translation editor: a segment-by-segment (CAT) view of a document: source on one side, your translation on the other, with TM matches, glossary hits, QA checks, and smart suggestions as you type.
• The Document editor: the same document rendered visually, so you review translations in the real layout: tables, headings, images, formatting.
• Language assets: the page where your glossaries, termbase, translation memory, and CAT settings live.

What the model and software each do

The model produces the target text using the selected language, domain, tone, and mode. Separate software handles exact Translation Memory reuse, glossary verification, locale-formatting rules, protected strings, and QA checks. These rule-based steps can be inspected and tested independently of the generated translation.

Where to go next

• New here? Start with the Quickstart: your first translation in about two minutes.
• Translating files? Read Document translation.
• Building a review workflow? Start with the Translation editor and Translation memory.
• Integrating from your own systems? See the API overview.