| 语音识别系统中多种特征参数组合的抗噪性 |
| |
| 引用本文： | 徐永华,孙炯宁. 语音识别系统中多种特征参数组合的抗噪性[J]. 金陵科技学院学报, 2006, 22(1): 35-37 |
| |
| 作者姓名： | 徐永华 孙炯宁 |
| |
| 作者单位： | 金陵科技学院信息技术学院,江苏,南京,210001 |
| |
| 摘 要： | 构建了基于连续隐马尔可夫模型(CHMM)的汉语数字串语音识别系统,为了提高系统在噪音环境下的鲁棒性,抑制平稳噪声及去除信道卷积噪声的影响,引入了动态参数,实验仿真表明采用MFCC参数及一阶、二阶差分及倒谱化明显提高了噪声环境下语音识别系统的识别性能。
|
| 关 键 词： | 动态特征参数 连续隐马尔可夫 语音识别 抗噪性 |
| 文章编号： | 1672-755X(2006)01-0035-03 |
| 修稿时间： | 2005-09-19 |
Anti-noise Research of the Various Feature Coefficient on the Speech Recognition System |
| |
| XU Yong-hua,SUN Jiong-ning. Anti-noise Research of the Various Feature Coefficient on the Speech Recognition System[J]. Journal of Jinling Institute of Technology, 2006, 22(1): 35-37 |
| |
| Authors: | XU Yong-hua SUN Jiong-ning |
| |
| Abstract: | Mandarin digital string speech recognition system is designed.Dynamic feature coefficient is introduced in order to enhance noise robust of the recognition system,reduce calm noise and decrease influence from the channel cumulus.The experiment results show that speech recognition rate in noise distinctly improves when choosing MFCC coefficient,one order,second order and cepstrum. |
| |
| Keywords: | dynamic feature coefficient CHMM speech recognition anti-noise |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
