| 基于小波变换和归一化自相关的基音检测算法 |
| |
| 引用本文： | 许钢,黄冰.基于小波变换和归一化自相关的基音检测算法[J].桂林电子科技大学学报,2003,23(6):1-5. |
| |
| 作者姓名： | 许钢 黄冰 |
| |
| 作者单位： | 桂林电子工业学院,通信与信息工程系,广西,桂林,541004 |
| |
| 摘 要： | 通过对小波变换的多分辨率分析和抗噪特性的研究,提出了一种基于小波变换和归一化自相关的基音周期检测算法。该算法能精确地从含噪语音中分离出包含基音信息的低频子带信号,并且对基音周期的突变能够很好地纠正。实验表明:该算法要优于现有的自相关基音检测算法,具有较高的精度、可靠性和抗噪性。
|
| 关 键 词： | 基音检测 小波变换 自相关函数 |
| 文章编号： | 1001-7437(2003)06-01-05 |
| 修稿时间： | 2003年9月3日 |
A Pitch Detection Algorithm Based on Wavelet Transform and Normalized Autocorrelation |
| |
| XU Gang,HUANG Bing.A Pitch Detection Algorithm Based on Wavelet Transform and Normalized Autocorrelation[J].Journal of Guilin Institute of Electronic Technology,2003,23(6):1-5. |
| |
| Authors: | XU Gang HUANG Bing |
| |
| Abstract: | This paper discusses the multiresolution performance and noise attenuation performance of wavelet transform. So a pitch detection algorithm based on wavelet transform and normalized autocorrelation is proposed. Using this algorithm, low frequency band signals, which include pitch information, can be precisely separated from noisy speech signal, and the pitch's saltation can also be corrected. The experimental results show that the algorithm has not only high reliability and precision, but also noise robustness. |
| |
| Keywords: | pitch detection wavelet transform autocorrelation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
