PHONEME RECOGNIZER BASED ON LONG TEMPORAL CONTEXT
The recognizer was developed at Brno University of Technology, Faculty of Information Technology. It can work in two modes: in offline mode - the input speech is read from input files and recognized phonemes are saved to output files or showed on a display; and in online mode - input speech is read from a device (microphone) and immediately showed on display or saved to a file. The recognizer has been successfully compiled under Linux (GCC) and under Windows (CBuilder) operation systems.

This packaging depends on the ATLAS linear algebra package.