汉语语音识别中融合发音信息的随机段模型研究  

Integrating articulatory information into stochastic segment models for continuous mandarin speech recognition

作　　者：晁浩[1] 刘志中[1] 薛霄[1] 

机构地区：[1]河南理工大学计算机科学与技术学院,河南焦作454000

出　　处：《计算机应用研究》2015年第4期1087-1090,共4页Application Research of Computers

基　　金：国家自然科学基金资助项目(91120303;90820303;90820011);河南省基础与前沿技术研究计划资助项目(132300410332)

摘　　要：提出了一种基于随机段模型的发音信息集成方法。根据随机段模型的模型特性,建立了阶层式人工神经网络来获取语音段信号属于各类音素的后验概率,并通过一遍解码的方式集成到随机段模型系统中。在"863-test"测试集上进行的汉语连续语音识别实验显示汉语字的相对错误率下降了5.93%。实验结果表明了将发音信息应用到随机段模型的可行性。This paper proposed a framework which attempted to incorporate articulatory information into the stochastic segment model based mandarin speech recognition system. According to the characteristics of the stochastic segment model,it used hierarchical artificial neural network to obtain the posteriors of speech signal belonging to the phonemes. Then,the posteriors were integrated into the stochastic segment model system in the first search process. Experiments conducted on＂863-test＂set show that about 5. 93% relative improvement can be achieved in the recognition accuracy. Thus,potential of the method is demonstrated.

关 键 词：语音识别 随机段模型 发音信息 阶层式人工神经网路 发音特征 

分 类 号：TP391.42[自动化与计算机技术—计算机应用技术]