Neural Network Language Model for Chinese Pinyin Input Method Engine.
Shenyuan Chen, Hai Zhao, Rui Wang · Institutional Repositories DataBase (IRDB) · 2015
Neural network language models (NNLMs) have been shown to outperform traditional n-gram language model. However, too high computational cost of NNLMs becomes the main obstacle of directly integrating it into pinyin IME that normally requires a real-time response. In this paper, an efficient solution is proposed by converting NNLMs into back-off n-gram language models, and we integrate the converted NNLM into pinyin IME. Our exper-imental results show that the proposed method gives better decoding predictive performance for pinyin IME with satisfied efficiency. 1