speech recognition engine

 Julius is a high-performance, two-pass large vocabulary continuous
 speech recognition (LVCSR) engine.
 .
 It supports N-gram based dictation, DFA grammar based parsing, and
 one-pass isolated word recognition. Phone context dependencies are
 supported up to triphone. It can perform a multi-model decoding, a
 recognition using several LMs and AMs simultaneously with a single
 processor, and also supports "hot plugging" of arbitrary modules at
 run time.
 .
 This package contains the executable applications, a set of tools
 useful to build recognition grammar and some examples (like a
 script to use Julius to execute some predetermined commands).