An improved Mandarin speech dictation machine for intelligent terminals of advanced communication networks

摘要

An improved experimental Mandarin dictation machine for the input of Chinese characters to computers and networks is described. Considering the special characteristics of the Chinese language, syllables are chosen as the basic units for dictation. The machine is divided into two subsystems. The first recognized the syllables using hidden Markov models (HMMs) trained by special algorithms. Because every syllable can represent many different characters with completely different meanings and can possibly form different multisyllabic words with syllables on its right or left, the second subsystem then identifies the exact characters from the syllables and corrects the errors in syllable recognition using a specially designed Chinese language model.