The IRST Language Modeling Toolkit features algorithms and data structures suitable to estimate, store, and access very large LMs. Our software has been integrated into a popular open source Statistical Machine Translation decoder called Moses, and is compatible with language models created with other tools, such as the SRILM Tooolkit.
We have migrated the IRSTLM website under SourceForge. Please, visit http://sourceforge.net/projects/irstlm to get in touch with IRSTLM. You can access the on-line documentation, the source code, and the latest release of the toolkit.
- June 3rd, 2013: the IRSTLM repository has been moved by SourceForge team; please update the link: https://svn.code.sf.net/p/irstlm/code
- April 29th, 2013: the new release 5.80.03 is now available in the sourceforge website
- August 17th, 2012: the new release 5.80.01 is now available in the sourceforge website
- November 18th, 2011: the new release 5.70.04 is now available in the sourceforge website
- October 5th, 2011: the new release 5.70.03 is now available in the sourceforge website
- September 26th, 2011: the new release 5.70.01 is now available in the sourceforge website
- April 15th, 2011: The SVN repository under Sourceforge is available again! at this address: https://irstlm.svn.sourceforge.net/svnroot/irstlm
Examples of usage can be found here
For the sake of development, some regression tests are available in the release. To performed the tests, additional data are needed, which is not distributed in the archive. These data can be accessed from here
address : user-irstlm AT list DOT fbk DOT eu
Marcello Federico, Nicola Bertoldi