| 汉语方言自动辨识系统中的语言建模 |
| |
| 引用本文： | 杨焕峥.汉语方言自动辨识系统中的语言建模[J].宁波职业技术学院学报,2007,11(5):71-73. |
| |
| 作者姓名： | 杨焕峥 |
| |
| 作者单位： | 无锡商业职业技术学院,江苏,无锡,214153 |
| |
| 摘 要： | 研究了在汉语方言辨识系统中如何建立合适的语言模型用以提高系统的辨识率。采用了基于统计的语言建模方法,首先把各方言的语音材料转写成音素序列,然后让机器统计方言音素的组配规则,建立各方言的语言模型。结果表明,加入语言模型的辨识系统比仅仅依靠声学模型进行辨识的系统,辨识效果要好得多。
|
| 关 键 词： | 方言 语音识别 音素 |
| 文章编号： | 1671-2153（2007）05-0071-03 |
| 收稿时间： | 2007-06-15 |
| 修稿时间： | 2007年6月15日 |
Research of language model in automatic chinese dialect identification system |
| |
| YANG Huan-zheng.Research of language model in automatic chinese dialect identification system[J].Journal of Ningbo Polytechnic,2007,11(5):71-73. |
| |
| Authors: | YANG Huan-zheng |
| |
| Abstract: | How to set up a portable language model which is used to improve the recognition rate in automatic Chinese dialect identification system is a critical part we look on.In this paper,we use a language model based on statistics.Firstly,we transform the speech material into phoneme sequence,and then let computer statistic the rule of all dialect phoneme's combination,set up language models of all Chinese dialect,and in doing so,we can identify dialects with the language models.The results of experiments show that the effect of the system used language model is better than the system using pure acoustic model. |
| |
| Keywords: | dialect speech recognition phoneme |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
