| 基于HMM和小波网络模型的抗噪语音识别方法 |
| |
| 引用本文： | 林遂芳,潘永湘,孙旭霞.基于HMM和小波网络模型的抗噪语音识别方法[J].系统仿真学报,2005,17(7):1720-1723. |
| |
| 作者姓名： | 林遂芳 潘永湘 孙旭霞 |
| |
| 作者单位： | 西安理工大学自动化与信息工程学院,陕西西安,710048 |
| |
| 摘 要： | 提出一种隐马尔可夫模型(HMM)和小波神经网络(WNN)混合模型的抗噪语音识别方法。该方法首先利用HMM对语音信号进行时序建模，并计算出待识语音对HMM的输出概率评分，再将此概率评分作为小波神经网络的输入，获取分类识别信息，最后根据混合模型的识别算法作出识别决策。实验结果表明，在噪声环境下，由于HMM的强时序建模能力和小波神经网络的强模式分类能力，该混合模型比单纯HMM具有更强的噪声鲁棒性，明显改善了语音识别系统的性能。
|
| 关 键 词： | 语音识别 隐马尔可夫模型（HMM） 小波神经网络（WNN） 噪声环境 |
| 文章编号： | 1004-731X(2005)07-1720-04 |
| 修稿时间： | 2004年4月15日 |
Noisy Speech Recognition Based on Hybrid Model of Hidden Markov Models and Wavelet Neural Network |
| |
| LIN Sui-fang,PAN Yong-xiang,SUN Xu-xia.Noisy Speech Recognition Based on Hybrid Model of Hidden Markov Models and Wavelet Neural Network[J].Journal of System Simulation,2005,17(7):1720-1723. |
| |
| Authors: | LIN Sui-fang PAN Yong-xiang SUN Xu-xia |
| |
| Abstract: | A new method for noisy speech recognition based on a hybrid model of hidden Markov models (HMM) and wavelet neural network (WNN) is presented. The HMM was employed to compute the Viterbi output score. Then the score was used as the input of WNN to acquire the classification information. The result of recognition was made by these two kinds of recognition information. Recognition experiment shows that this hybrid model has higher performance than hidden Markov model in noisy speech recognition. |
| |
| Keywords: | speech recognition hidden markov model wavelet neural networks noisy environments |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
