• INTRODUCTION
• INTRODUCTION
• ABOUT THIS MANUAL
• OPEN TRANSLATION
• WHY TRANSLATE
• WHAT IS OPEN TRANSLATION?
• CURRENT STATE
• BASIC CONCEPTS
• TRANSLATION
• LOCALISATION
• INTERPRETING
• DICTIONARIES
• TRANSLATION MEMORY
• MACHINE TRANSLATION
• STANDARDS
• TECHNICAL CONCEPTS
• TRANSLATION PROCESSES
• TRANSLATION
• WORKFLOW
• THE TRANSLATION INDUSTRY
• OPEN TRANSLATION
• COMMUNITY
• ROLES IN OPEN TRANSLATION
• COMMUNITY MANAGEMENT
• QUALITY CONTROL
• QUALITY CONTROL
• REPUTATION METRICS
• CASE STUDIES
• GLOBAL VOICES
• WIKIPEDIA
• FLOSS MANUALS
• OLPC & SUGAR
• YEEYAN
• TRANSLATING TEXT
• INTRODUCTION
• WEB TRANSLATION SYSTEMS
• PREPARING CONTENT
• TRANSLATION TIPS
• CMS
• WIKIS
• TRANSLATING VIDEO
• SUBTITLES
• FILE FORMATS
• FINDING
• CREATING
• PLAYING
• DISTRIBUTION
• TRANSLATING IMAGES
• INTRODUCTION
• FORMATS
• TOOLS
• SVG
• TRANSLATING SVG
• SCRIPTING SVG
• INTELLECTUAL PROPERTY
• FREE SOFTWARE
• LICENSING
• MACHINE TRANSLATION AND COPYRIGHT
• OPEN SOURCE TOOLS
• ANAPHRASEUS
• APERTIUM
• GAUPOL
• GLOSSMASTER
• MOSES
• OKAPI FRAMEWORK
• OMEGAT+
• OMEGAT
• OPENOFFICE
• POOTLE
• TRANSLATE TOOLKIT
• VIRTAAL
• WWL
• ADVANCED TECH ISSUES
• TRANSLITERATION
• MACHINE TRANSLATION
• TRANSLATION MEMORY
• CHARACTER ENCODING
• WEB FONTS
• SCORING
• APPENDICES
• GLOSSARY
• CREDITS

Apertium

Apertium is a software tool for translating one language into a closely-related other through the use of a computer program rather than human interpretation.  Apertium uses a language-independent translation work flow, has tools to manage the linguistic data surrounding the two languages (a "language pair") and already contains linguistic data for many language pairs which are encoded in an XML-based format. 

What's it Used for?

Apertium is used mainly for translating closely-related language pairs.  Originally produced as part of the OpenTran project funded by the Spanish government, Apertium's strength is its strong translation support for languages closely related to Spanish. For example, Spanish-Portuguese or Spanish-Catalan.  Its focus on other languages has been limited due to its history.  However, because it is open source and has such a strong community, users have been able to push Apertium into more dissimilar language pairs with the hope that the platform will be able to go much further than languages similar to Spanish.

With Apertium, users are able to create machine translation systems for different languages by cataloging the linguistic data in a specific XML format that Apertium can then interpret.  That interpretation is through a process somewhat like an assembly line with each piece of language going through the following steps:

1. De-Formatting:  stripping the formatting (e.g. HTML or RDF) from the text
2. Morphological Analysis:  assigning grammatical designations for the text
3. Part-of-Speech Disambiguation:  determining parts-of-speech for uncertain terms
4. Shallow Structural Transfer:  processing phrases for differences in linguistic grammar
5. Lexical Transfer:  assigning the right linguistic term from the other language
6. Morphological Generation:  assigning grammatical designations for the new text
7. Re-Formatting:  restoring the original formatting to the new text 

Resources

There are many resources for the Apertium platform that can help out the new user:

• Apertium Home Page: http://www.apertium.org/
• Apertium Wiki: http://wiki.apertium.org/wiki/Main_Page
  
• Apertium Mailing List: https://lists.sourceforge.net/lists/listinfo/apertium-stuff
• Apertium IRC (Chat): http://xixona.dlsi.ua.es/cgi-bin/cgiirc/irc.cgi
• Apertium News RSS: http://sourceforge.net/export/rss2_projnews.php?group_id=143781&rss_fulltext=1

Apertium is Free Software licensed under the GPL.