CMU Sphinx Downloads

CMU Sphinx toolkit has a number of packages for different tasks and applications. It's sometimes confusing what to choose. To cleanup, here is the list

• Pocketsphinx — recognizer library written in C.
• Sphinxtrain — acoustic model training tools
• Sphinxbase — support library required by Pocketsphinx and Sphinxtrain

• Sphinx4 — adjustable, modifiable recognizer written in Java
• CMUclmtk — language model tools

We recommend you to use the latest available releases:

• sphinxbase-0.8
• pocketsphinx-0.8
• sphinxtrain-1.0.8

• sphinx4-1.0beta6
• cmuclmtk-0.7

If you want to try bleeding edge version, checkout from subversion. Then compile packages from the source code, but remember that there is no guarantee they will be stable:

http://sourceforge.net/p/cmusphinx/code/

Older releases and files could be found on SourceForge http://sourceforge.net/projects/cmusphinx/files/

We do not maintain distribution-specific packages yet, but help to update them is truely appreciated. Some distributions already include CMUSphinx packages:

On Fedora version 17 and later pocketsphinx is a part of “everything” repository. Try:

sudo yum install pocketsphinx

CMUSphinx assumes that you use the statistical models which describe language. There are many models trained for various acoustic conditions and various performance requirements. We collect the best models available at our download page. We hope you'll be able to find the best model for your language there:

Download Models

Most of the models are pretty large since they are trained on the large amount of data and describe the complex language. We mostly distribute them through Bittorent

To download models though our torrent tracker or to help to distribute them please visit the tracker web page:

http://cmusphinx.info

If you are willing to share your own data through this tracker, please contact mailto:cmusphinx-devel@lists.sourceforge.net