TODAY’S PATENT – SPEECH RECOGNITION APPARATUS
This invention was invented jointly by Yasuhiko Kato, Masao Watari, Miyuki Tanaka, Akihiko Fujinaka which was later assigned to Sony corporation on 10th November 1992. This invention was granted the utility patent on 14th February 1995 by the USPTO bearing the patent grant number. US5390280A. this invention is a Fourier converting section.
The Fourier converts a noise signal supplied from a microphone via an analogue to digital converter to calculate the noise signal’s power spectrum. A system controller compares the current average value of the power spectrum of the signal being outputted from the microphone to the current average value of the power spectrum of a noise signal stored in a noise memory. it sends a controlling signal to a sound storing and instructing it to store the signal being outputted from the microphone at the time into the noise memory. As a result, the signal currently being out casted from the microphone is stored into the noise memory in place of the noise signal currently being stored in the noise signal this helps in recognition of the identical voice of the commander or the user.