A real-time 64-monosyllable recognition LSI with learning mechanism

In the paper, a real-time 64-mono-syllable recognition LSI is presented. The LSI accepts 11.6 ms speech frame and outputs a 6-bit symbol-code for each frame by the end of the next frame in a pipelining manner. The recognition method is based on the Hidden Markov Model (HMM) and is speaker-independent. An on-chip learning mechanism has also been designed, but the circuit is off-chip for the present implementation because of the restriction of LSI area. The LSI is fabricated by VDEC Rohm with a 0.6 /spl mu/m CMOS process on a 4.5 mm/spl times/4.5 mm chip.