|
Large Vocabulary Conversational Speech Recognition (LVCSR)
is an emerging technology that draws heavily from the diverse research
areas of statistical pattern recognition, digital signal processing,
artificial intelligence, linguistics, and information theory. ISIP has
embarked on an ambitious project to develop a public domain LVCSR
system. Their research focus is currently in building an efficient
decoder capable of handling VCSR cross-word triphones, and generation
of word-graphs starting from N-Gram language models. This system also
has HMM training and front-end modules incorporated in it.
|