Tibetan Syllable Component Recognization with Memristor

This paper highlights a Tibetan syllable component recognization method by using memristors. Firstly, a memristor crossbar array with an external control circuit is designed to present both elevated device density and the potential for parallel computing. Then a syllable component recognization method is implemented to classify syllable into seven types, prefix letter, root letter, superscript letter, subscript letter, vowel letter, suffix letter, and farther suffix letter, matching Tibetan characters by computing total current value in memristor crossbar array. Finally, Tibetan characters is re-encoded in binary according to the last two digits of their Unicode, and input by controlling the voltage in memristor crossbar array. Experiments demonstrate that our method recognizes general Tibetan syllable components.