A Mandarin speech dictation system based on neural network and language processing model

摘要

A user friendly Mandarin speech dictation system based on a neural network and language model is proposed. The proposed network is composed of a dynamic programming model and a Bayesian neural network which can treat the problem of time axis distortion and spectral pattern variation respectively. Besides, a statistical language model with hierarchical syntactical analysis is proposed to map the syllables into Chinese characters. The overall recognition rate of the system can achieve about 88.2%.