speech recognition(ASR) is a technology which enables machines to recognize
and convert speech signal to corresponding texts or commands.
In 1997, dictation machine based on ASR is selected by US
press as one of 10 events in
computer science. Fields related to ASR include signal processing,
probability theory, information theory, acoustic mechanism,
hearing mechanism, artificial intelligence, etc.
focus on large vocabulary continuous speech
recognition(LVCSR) for natural speech, converting speech
data in broadcast, telecast, telephone
dialogues into text information.