语音信号增强算法研究与实现

Research and Implementation on the Algorithm for Speech Enhancement

研究基于语音短时幅度谱估计的语音增强方法,主要包括振幅谱相减、功率谱相减以及改进算法,给出了上述语音增强方法及其改进形式的基本原理、实现方法,以及它们在各种信噪比下的语音增强效果,比较和揭示了这些方法之间的区别和联系。与常用谱相减算法做对比,改进算法能有效消除"音乐噪声",在很大程度上提高了信噪比。实验表明,在加性平稳噪声(白噪声和有色噪声)条件下,该语音增强方法具有语音增强效果好、适应信噪比范围大、方法简单、易于实时处理等优点。

Speech processing systems are inevitably interfered by various noises. The noise may degrade the quality and the intelligibility of the processing systems. Seriously,the system couldn't work well. To minimize the effects of the noise on the performance of the processing systems,speech enhancement technology is applied in the various speech processing systems. Consequently,the study of speech enhancement technology is very significant. In this thesis,we study some speech enhancement methods based on short-time spectral magnitude of noisy speech,such as spectral magnitude subtraction,power spectrum subtraction and some improving methods. We study the principles,the implementation of these methods including their improved forms,and their performance under the wide SNR( signal-to-noise ratio) range. Meanwhile we reveal the differences and the relations among these speech enhancement methods. Under this condition,the noise( white or color) is additive and stationary. The experiment results indicate that methods can greatly improve the quality and intelligibility of noisy speech. And they have widely applicable SNR range and less computation load. Our work in this thesis can be the basis of further development in the practical speech enhancement systems.