Fan Zhang, Junwei Cao, et al.
IEEE TETC
In this paper we describe a. method for recognizing command words in a real-time speaker dependent isolated word speech recognition system such as the one developed at IBM. In such a system a few words are designated as commands and these may occur interspersed with other words in dictation. Given a set of command word; and a. few training tokens of each of the command words we construct special models for each of the command words and use a key-word spotting algorithm to identify occurrences of any of the command words very accurately. We are able to achieve near 98% detection rate with negligible false alarm rates.
Fan Zhang, Junwei Cao, et al.
IEEE TETC
Rajeev Gupta, Shourya Roy, et al.
ICAC 2006
David S. Kung
DAC 1998
Kafai Lai, Alan E. Rosenbluth, et al.
SPIE Advanced Lithography 2007