Jump to content

OmegaT

From Alfawiki
Adapted from the English Wikipedia article OmegaT, available under CC BY-SA 4.0Updated 9 August 2025
OmegaT
Company typeFree software
IndustryComputer-assisted translation software
Founded2000
FounderKeith Godfrey
Key people
Aaron Madlon-Kay (head of development team)
ProductsComputer-assisted translation application
Websitehttps://omegat.org/

OmegaT is a computer-assisted translation application developed in the Java environment. Released as free software, it was initially created by Keith Godfrey in 2000 and is currently overseen by a development team headed by Aaron Madlon-Kay.

Tailored for professional translators, OmegaT includes functionalities such as customizable segmentation using regular expressions, translation memory featuring fuzzy matching and propagation, glossary and dictionary integration, reference material lookup, and inline spell checking powered by Hunspell dictionaries.

Compatible with Linux, macOS, Microsoft Windows, and Solaris, OmegaT requires Java 8 to operate. It supports 27 different languages. According to a 2010 survey involving 458 professional translators, OmegaT's usage was approximately one-third that of rivals like Wordfast, Déjà Vu, and MemoQ, and about one-eighth compared to the leading software, Trados.

History

Keith Godfrey originally developed OmegaT in 2000 using C++. The initial public release appeared in February 2001 after being rewritten in Java, utilizing a proprietary translation memory format. Early editions handled unformatted text and HTML files with segmentation based on blocks (paragraphs) rather than individual sentences.

Development and software releases

Under the guidance of Aaron Madlon-Kay, OmegaT's development takes place on SourceForge. The project delivers frequent updates—generally two to three bug fixes or feature improvements per release—including a "standard" edition with comprehensive documentation and a "latest" edition that introduces new features without full documentation. The source code is openly available through the SourceForge repository.

How OmegaT works

OmegaT organizes translation work into projects structured as folders with designated names. Translators place source documents awaiting translation into the /source/ directory or its subdirectories. The Editor pane presents these documents segmented for translation one piece at a time. When prompted, OmegaT produces translated files within the /target/ folder.

Additional important directories include /tm/ for existing translation pairs stored in .tmx format, /tm/auto/ which holds automatic translations of exact matches, /glossary/ for glossary files, and /dictionary/ containing StarDict and .tbx dictionary files.

During segment translation, OmegaT automatically scans the /tm/ directory for similar previously translated segments, displaying these in the Fuzzy Matches pane for easy insertion. The Glossary and Dictionary panes provide automatic lookups from their respective directories. An optional Machine Translation pane can show outputs from services such as Google Translate.

When leaving a segment, OmegaT records the source-target pair into its memory database and saves it to disk in .tmx format for future reuse across various projects and tools. Version 3.1 introduced an option to prevent targets identical to their sources from being saved automatically, with a shortcut available to override this behavior—for instance, when translating numbers or source code.

Translators can generate partially completed translations of source files at any stage. Untranslated segments are copied exactly as-is; however, it is advisable to run the Validate command beforehand to detect tag inconsistencies. Version 3.1 also added a command (with shortcut) that restricts operations to the current file, facilitating partial deliveries or rapid updates.

OmegaT Capabilities

OmegaT encompasses a broad range of features typically found in commercial CAT software, including the creation, import, and export of translation memories; fuzzy match functionality; glossary consultation; as well as reference and concordance searches.

Additionally, it delivers several distinctive functions not universally available in other CAT tools:

  • An integrated "Instant Start" guide designed for rapid user familiarization.
  • Ability to translate numerous files concurrently in diverse formats within subdirectories, utilizing multiple translation memories, glossaries, and dictionaries—limited solely by system memory capacity.
  • User-definable file extensions and character encodings; selective translation of specific document components (such as bookmarks in OpenOffice.org Writer, footnotes in Microsoft Office 2007/2010, and ALT text in HTML); plus options to manage non-standard elements from external memories.
  • Customizable segmentation criteria based on regular expressions, adaptable according to language or file type.
  • Enhanced editing capabilities including direct navigation to untranslated segments; undo/redo functions; clipboard operations; case toggling; viewing source text for translated segments; inline spell-checking powered by Hunspell dictionaries; and interactive spell-check accessible via mouse clicks.
  • Convenient insertion of fuzzy matches through keyboard shortcuts or mouse input, featuring color-coded similarity levels; display of translator details such as name, date, and time per segment; glossary entries insertable via mouse; options to copy source text into target fields or automatically apply the best fuzzy match.
  • Robust search functionality covering source text, target text, additional memories, and reference files with support for case sensitivity and regular expressions; double-clicking search results opens the corresponding segment directly.
  • Validation of tags after translation to prevent errors; project statistics providing progress updates before or during the translation process.
  • Integration with machine translation engines like Apertium, Belazar, Deepl, and Google Translate displayed in a separate window.
  • Highly flexible user interface panels that can be repositioned, maximized, tiled, tabbed, or minimized according to user preference.

Supported Document Formats

OmegaT supports direct translation of numerous file types identified by their extensions. Users have the option to modify how extensions are treated and adjust preferred encodings to override default settings. Formatting within documents is handled by converting formatting elements into tags, following a method similar to that employed by commercial CAT tools.

Natively Supported Formats

Formats supported indirectly

OmegaT extends its compatibility to additional file types by either treating their extensions as plain text or by transforming them into formats it natively supports.

XLIFF compatibility

The Rainbow utility from the Okapi Framework converts multiple file types into XLIFF format, making them usable within OmegaT. It can also produce fully structured OmegaT project directories to facilitate efficient translation workflows.

Gettext PO format support

Conversion tools such as Debian’s po4a transform formats like LaTeX, TeX, and POD into Gettext PO files suitable for OmegaT. Additionally, the Translate Toolkit enables conversion of Mozilla properties files, DTDs, CSV files, select Qt TS files, and certain XLIFF files into Gettext PO format.

Office Open XML and ODF compatibility

Documents created in Microsoft Word, Excel, and PowerPoint from versions 97 through 2003 can be converted to more modern formats such as Office Open XML (used in Microsoft Office 2007/2010) or OpenDocument Format (ODF) utilized by OpenOffice.org. Some formatting may be lost during this conversion process.

Handling of Trados® .ttx files

Support for Trados® .ttx files is provided through the Okapi TTX Filter.

TMX translation memories

While OmegaT’s internal translation memory operates behind the scenes, it exports autosaved translation units into three external TMX files: the native OmegaT TMX, level 1 TMX, and level 2 TMX.

  • Native TMX: designed specifically for OmegaT projects.
  • Level 1 TMX: retains textual content and is compatible with CAT tools supporting TMX levels 1 and 2.
  • Level 2 TMX: preserves both text and inline tags; compatible with TMX level 2 CAT applications.

Exports at level 2 embed OmegaT’s internal tags within TMX tags, allowing matches to be recognized by TMX level 2 tools such as Trados and SDLX. OmegaT can import TMX files up to version 1.4b level 2, preserving tag structures and ensuring interoperability with software like Transit.

Glossary files

The primary glossary format for OmegaT consists of tab-separated UTF-8 encoded plain text files (.txt) that follow a simple layout: source term, target term(s), and an optional comments column. These glossaries can be created using any text editor. OmegaT also supports standard CSV and TBX glossary formats.

The OmegaT initiative

As an open-source project, OmegaT depends heavily on volunteer contributions. Although programming plays a vital role, assistance in all areas is encouraged. Users are also empowered to tailor OmegaT according to their individual requirements.

Localization efforts

Volunteers have translated OmegaT’s user interface and documentation into approximately thirty languages. Translators can contribute to the UI elements, the "Instant Start" tutorial, or the comprehensive user manual. All completed translations are incorporated into the official release packages.

User-developed tools

The OmegaT user base frequently fills functionality gaps by developing macros, scripts, and auxiliary programs that enhance the software’s capabilities. Some of these community-created tools have eventually been integrated into OmegaT itself. For instance, when only paragraph segmentation was available natively, users produced OpenOffice.org macros for sentence-level segmentation. Similarly, before automated translation memory leveraging existed in OmegaT, a script for merging TMX files was developed by users. Spell-checking features were also initially implemented through user solutions prior to their inclusion in the core application.

OmegaT within DGT

Most recent update: 2021-03-21

The Directorate-General for Translation (DGT) of the European Commission utilizes OmegaT in combination with proprietary CAT tools. DGT manages a personalized version called DGT-OmegaT, enhanced with functionalities designed to address their particular requirements. Furthermore, it produces auxiliary tools—such as a project automation Wizard, Tagwipe for removing docx tags, and TeamBase for collaborative memory sharing in real time—distributed as free and open-source software.

Benten

Most recent update: 2018-04-07

Benten is an XLIFF editor based on the Eclipse platform that incorporates OmegaT’s codebase for translation memory matching. Its development receives partial financial support from the Japanese government.

Autshumato translation suite

Most recent update: 2017-02-28

Autshumato is a comprehensive translation suite featuring a CAT tool developed on OmegaT, along with an aligner, PDF extraction utility, TMX editor, and a publicly accessible translation memory compiled from web-crawled content. Future enhancements include integrating a terminology management system and machine translation capabilities. This initiative is sponsored by South Africa’s Department of Arts and Culture and depends on OpenOffice.org.

OmegaT+

Most recent update: 2012-10-24

OmegaT+ originated as a fork from OmegaT version 1.4.5 (2005) that functions similarly but incorporates unique features; projects created with OmegaT+ are not compatible with standard OmegaT.

Boltran

Most recent update: 2010-10-12

Boltran is a web-based application modeled after OmegaT’s workflow, constructed using its source code, and supports exporting projects in the OmegaT format.

See also

External links

User support

Related profiles

OOpen Language ToolsGGlobalSightOOpenTMSEEucalyptus (software)
Retrieved from "https://www.alfawiki.org/index.php?title=OmegaT&oldid=25989"