Recurrent Neural Network Based Language Model Adaptation for Accent Mandarin Speech

摘要：

　　In this paper,we propose to adapt the recurrent neural network(RNN)based language model to improve the performance of multi-accent Mandarin speech recognition.N-gram based language model can be easily applied to speech recognition system,but it is hard to describe the long span information in a sentence and arises a serious phenomenon of data sparsity.Instead,RNN based language model can overcome these two shortcomings,but it will take a long time to decode directly.Taking these into consideration,this paper proposes a method which combines these two types of language model(LM)together and adapts the RNN based language model to rescore lattices for different accents of Mandarin speech.The architecture of the adapted RNN LM is accent-specific top layers and shared hidden layer.The accent-specific top layers are used to adapt different accents and the shared hidden layer stores history information,which can be seen as a memory layer.Experiments on the RASC863 corpus show that the proposed method can improve the performance of accented Mandarin speech recognition over the baseline system.

关键词： Multi-accent Speech recognition RNN language model Adaptation

作者: Hao Ni Jiangyan Yi Zhengqi Wen Jianhua Tao

作者单位: National Laboratory of Pattern Recognition,Institute of Automation,Chinese Academy of Sciences,Beiji National Laboratory of Pattern Recognition,Institute of Automation,Chinese Academy of Sciences,Beiji

会议类型: 国际会议

会议名称: 第七届全国模式识别学术会议(The 7th Chinese Conference on Pattern Recognition,CCPR2016)

会议地点: 成都

会议语种:英文

页码: 607-617

在线出版日期: 2016-11-03（万方平台首次上网日期，不代表论文的发表时间）