计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (10): 127-130.

• 数据库、信号与信息处理 • 上一篇    下一篇

触发式语言模型下的混淆网络解码方法

杨春风1,王欢良1,2   

  1. 1.西北师范大学 数学与信息科学学院,兰州 730070
    2.哈尔滨工业大学 计算机科学与技术学院,哈尔滨 150001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-04-01 发布日期:2011-04-01

Decoding method integrating of confusion network based on Trigger language model

YANG Chunfeng1,WANG Huanliang1,2   

  1. 1.College of Mathematic and Information Science,Northwest Normal University,Lanzhou 730070,China
    2.School of Computer Science and Technology,Harbin Institute of Technology,Harbin 150001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-04-01 Published:2011-04-01

摘要: 将触发式语言模型应用于混淆网络解码过程来提高汉字识别率。为了利用词间的长距离依赖信息,提出了基于词义类对触发式语言模型的混淆网络解码方法。实验结果显示,该方法可以使汉字错误率相对下降7.9%。

关键词: 语音识别, 触发式语言模型, 混淆网络

Abstract: The decoding method integrating of confusion network is studied.Trigger language model based on semantic class pairs is proposed to model dependence relationship between long-span words.The model is integrated with confusion network decoding process.Different speech recognition systems utilize different knowledge sources and modeling methods,consequently their error pattern is also different.Experimental results show the method can relatively reduce character error rate by 7.9% respectively.

Key words: speech recognition, Trigger language model, confusion network