ofxASR is an openframeworks addon for speech recognition. It is under development and the first release version is planed to available on June 25, 2010.
Currently it use CMU Sphinx3(http://cmusphinx.sourceforge.net/) as its Automatic Speech Recognition(ASR) engine, but it also designed allowing to use other ASR engine, such as Microsoft’s SAPI or Mac OSX Speech API.
The developing of this project is a part of the project CCA(http://nuicode.com/projects/cca-alpha), but ofxASR is standalone, it can be applied on any Openframeworks based program which need a speech recognition feature.
This project is still under developing so we do not provide download now, however, its source codes is visible at any time for anyone.
The project hosting site: http://code.google.com/p/ofxasr, you can view the current source code on that. Any feedback is welcome.
Which language a ASR engine support depends on the type of the acoustic model and the language model, however, to train such models is not an easy thing, that need huge of sound data.
There’re some open source organisations provide models, such as VoxForge(http://www.voxforge.org/). Most open source models are in English, so, the first release of ofxASR shipped an opensource English model.
It is very easy to replace the model by yourself, just download(or train) a new model and copy it to the special path.