Since I'm doing some research on Machine Translation, I would like to share with all of you what I call a "Moses Starter Kit":
First of all you need to install the Moses Decoder, I recommend DoMY (Do Moses Yourself) because it's really easy to install and has a lot of potential with good documentation. (http://www.precisiontranslationtools...=26&Itemid=22, login needed).
Probably you will need tools to handle bilingual data (clean, filter, sort, etc) and to export from TMX or XLIFF to txt (DoMY only works with txt files to translate).
- Arpentium Tools: Apertium is a toolbox to build open-source shallow-transfer machine translation systems, especially suitable for related language pairs: it includes the engine, maintenance tools, and open linguistic data for several language pairs. I've downloaded only the tools: (svn co https://apertium.svn.sourceforge.net...pertium-tools/)
- OKAPI Framework: The Okapi Framework is a cross-platform and free open-source set of components and applications that offer extensive support for localizing and translating documentation and software. (Downloads - okapi - Cross-platform framework for localization and translation tools - Google Project Hosting)
- M4Loc: The goal of this project is to provide tools to translate localization-specific formats with Moses and to integrate Moses in localization workflows (http://code.google.com/p/m4loc/downloads/list). THIS TOOL REQUIRES OKAPI FRAMEWORK INSTALLED.
- LFAligner: LF Aligner helps translators create translation memories from texts and their translations. It relies on Hunalign for automatic sentence pairing. Input: txt, doc, docx, rtf, pdf, html. Output: tab delimited txt, TMX and xls. With web features (LF Aligner - Browse Files at SourceForge.net)
My own "Moses Starter Kit" is running Ubuntu 10.04 on a VirtualBox Virtual Machine with 2 processors and 1gb of RAM. It tooks like 5 hours to train the ONU + EC corpus (3 n-grams EN-ES).
Hope this help and you are welcome to suggest other Machine Translation related tools.