Position: Index > Filter Circuit >

SPEECH_RECOGNITION_FILTER

2017-02-20 03:02  
Declaration:We aim to transmit more information by carrying articles . We will delete it soon, if we are involved in the problems of article content ,copyright or other problems.


SPEECH_RECOGNITION_FILTER
Voice signal picked up by microphone is preamplified and sent through 3-kHz low-pass passive filter C1-R1 to 1-kHz high-pass active filter and 1-kHz low-pass active filter using sections of LM3900 quad opamp. Diode symbols on opamps indicate use of current mirrors for noninverting inputs. Outputs are sampled about 60 times per second to implement speech recognition algorithm of computer, which counts number of high-pass and low-pass zero crossings per second and compares results with series of word models in memory to determine most likely match.-J. R. Boddie, Speech Recognition for a Personal Computer System, BYTE, July 1977, p 64-68 and 70-71.