| 一种基于听觉模型的抗噪语音识别特征提取方法 |
| |
| 引用本文： | 焦志平,张雪英,赵姝彦.一种基于听觉模型的抗噪语音识别特征提取方法[J].太原理工大学学报,2005,36(1):13-15,22. |
| |
| 作者姓名： | 焦志平 张雪英 赵姝彦 |
| |
| 作者单位： | 太原理工大学,信息工程学院,山西,太原,030024 |
| |
| 基金项目： | 国家自然科学基金资助项目(60472094)，教育部留学回国科研启动基金资助项目(教外司留[2004]176号文件)，山西省留学回国人员科研基金资助项目(200224) |
| |
| 摘 要： | 提出了一种新的语音识别特征提取方法。该方法是建立在听觉模型的基础上,通过计算语音的上升过零率作为频率信息并通过非线性幅度加权相结合来获取语音特征。仿真实现了中小词汇量、孤立词的语音识别,得到了较好的实验结果,证明了此方法具有较强的抗噪声性能。
|
| 关 键 词： | 语音识别 特征提取 听觉模型 |
| 文章编号： | 1007-9432(2005)01-0013-03 |
A Noise-Robust Feature Extraction Method Based on Auditory Model in Speech Recognition |
| |
| JIAO Zhi-ping,ZHANG Xue-ying,ZHAO SHu-yan.A Noise-Robust Feature Extraction Method Based on Auditory Model in Speech Recognition[J].Journal of Taiyuan University of Technology,2005,36(1):13-15,22. |
| |
| Authors: | JIAO Zhi-ping ZHANG Xue-ying ZHAO SHu-yan |
| |
| Abstract: | This paper introduces a new feature extraction method which is based on auditory model. The frequency information of speech signal is obtained by zero-crossing intervals, and amplitude information is incorporated by nonlinear transform. A medium and small vocabulary isolated word speech recognition system is simulated. The results show that this method has better robustness than traditional methods. |
| |
| Keywords: | speech recognition feature extraction auditory model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
