【摘要】 针对语音信号在传输和处理的过程中经常会受到不同程度噪声干扰的问题,在传统小波阈值去噪方法的基础上提出了一种改进型去噪方法。该方法采用遗传算法对小波阈值参数进行自动寻优,并对传统遗传算法加以改进,克服了其容易陷入局部最优的缺点,实现了阈值的自适应选取,解决了人为选择阈值导致去噪效果不明显的问题。实验表明,改进方法能够较好的去除语音信号中的噪声干扰,信噪比较高,相比于通用阈值法提高了15%。更多还原

【Abstract】 For the problem of speech signal always subject to kinds of noise during transmission,a kind of improved de- noising algorithm was put forward based on traditional wavelet threshold de- noising method. The method uses the Genetic algorithm to search the optimization of threshold parameters automatically,and overcome the shortcomings of its easy to fall into local optimum by improve the traditional genetic algorithm. The improved method realizes the adaptive threshold selection and solves the problem that de- noising effects are not obvious caused by the artificial selection of threshold parameters. The experimental results show that the improved method can reduce the noise effectively,gets a higher SNR,improved by 15% compared with the algorithm which use the default wavelet threshold.更多还原