===================================== Speech recognition software for human beings.
SpeechControl has its own website and git repository; feel free to check it out.
SpeechControl's the attempt of the Synthetic Intellect Institute to provide an efficient means of speech recognition technologies to the Linux desktop. SpeechControl is a speech recognition toolkit, allowing you to:
- train multiple acoustic models
- manipulate and redistribute sessions made with SpeechControl.
- upload session information to VoxForge
- dictate text to applications
- control your desktop using arbitrary commands
- provide speech recognition to other applications
For more detailed instructions look into the INSTALL file. Usual build procedure is as follows:
$ python daemonsetup.py build
$ python packagesetup.py build
$ python trainersetup.py build
# python daemonsetup.py install
# python packagesetup.py install
# python trainersetup.py install
First and foremost issue the following commands:
$ ./mk-scdirs
$ ./mk-psext
We use nosetests for launching tests - go to the directory with a subproject (daemon, for example) and run `nosetests' from there. You should see the report.
Because of the interactive nature of the PocketSphinx (it needs voice input from microphone) there are special tests (under tests/special) that you have to run yourself - they will usually need your voice input.
Use our GitHub Issues feature enabled for SpeechControl to post about problems.
You can reach the development team behind SpeechControl at their mailing list spchcntrl-devel@thesii.org. The people who helped produce and also use SpeechControl have a mailing list at spchcntrl-users@thesii.org. You can openly subscribe and add content to both of these mailing lists. Thank you for using our software. More information regarding SpeechControl can be found at http://thesii.org/wiki/SpeechControl.
One Vision. One Purpose
The Synthetic Intellect Institute :: F/OSS AI