| 采用DTW算法和语音增强的嵌入式声纹识别系统 |
| |
| 引用本文： | 周跃海,童峰,洪青阳. 采用DTW算法和语音增强的嵌入式声纹识别系统[J]. 厦门大学学报(自然科学版), 2012, 51(2): 174-178 |
| |
| 作者姓名： | 周跃海 童峰 洪青阳 |
| |
| 作者单位： | 1. 厦门大学海洋与环境学院,水声通信与海洋信息技术教育部重点实验室
2. 厦门大学信息科学与技术学院,福建厦门,361005 |
| |
| 基金项目： | 福建省高校产学合作重大项目(2010H6022);福建省杰出青年科学基金项目(2010D003);厦门市科技计划项目(3502Z20113008) |
| |
| 摘 要： | 动态时间规整(dynamic time warping,DTW)是一种相对简单成熟的算法,广泛用于语音识别系统中.针对环境噪声对声纹识别系统性能的影响,用信噪比关联谱减及自适应门限端点检测进行抗噪声处理,在此基础上采用DTW算法设计了基于嵌入式ARM9平台的声纹识别实现方案,并给出了带噪环境下的声纹识别实验结果.
|
| 关 键 词： | 声纹识别 动态时间规整 谱减 嵌入式系统 |
Embedded Voiceprint Recognition System Based on DTW Algorithm and Speech Enhancement |
| |
| ZHOU Yue-hai , TONG Feng , HONG Qing-yang. Embedded Voiceprint Recognition System Based on DTW Algorithm and Speech Enhancement[J]. Journal of Xiamen University(Natural Science), 2012, 51(2): 174-178 |
| |
| Authors: | ZHOU Yue-hai TONG Feng HONG Qing-yang |
| |
| Affiliation: | 1.Key Laboratory of Underwater Acoustic Communication and Marine Information Technology,Ministry of Education, College of Oceanography and Environmental of Science,Xiamen University,2.School of Information Science and Technology,Xiamen University,Xiamen 361005,China) |
| |
| Abstract: | Dynamic time warping(DTW) is a relatively simple and effective algorithm,which is widely used in speech recognition systems.In order to deal with the performance degration caused by background noise,this paper utilized signal to noise ratio(SNR) spectral substration algorithm and adaptive threshold endpoint detection algorithm.Basing on the two algorithms,the speech recognition system is successfully implemented on ARM9 embedded microprocessor.Finally the experimental results are given to verify the effectiveness of the proposed system. |
| |
| Keywords: | speech recognition dynamic time warping spectral subtraction embedded systems |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
