一种改进的小波变换基音检测算法

Improved arithmetic of wavelet transformation for pitch frequency detection

为了克服目前诸多基音频率检测算法精度低、复杂度高和鲁棒性差的缺点,提出一种利用小波变换结合线性预测的去噪方法,并通过自相关函数和平均幅度差函数的线性组合得到基音频率,实验验证,该方法性能明显优于自相关函数法和平均幅度差函数法,减少了倍频及半频的误差提取,提高基音频率的提取精度,同时在低信噪比的情况下,仍能精确提取出基音频率。

In order to overcome the disadvantages of low accuracy,high complexity and lack of robustness in many actual pitch frequency algorithms,this paper proposes the method that uses wavelet transformation with linear forecast for removing noise and the resonant influence.And an improved autocorrelation function and average range difference function method is used to test the peak position.The simulation experiment shows that the approach is superior to the method of autocorrelation function and the method of average magnitude difference function.Not only does the method reduce the frequency and half frequency of times error,improving the frequency to extract the extraction of pitch precision,but also is under the lower circumstance signal-to-noise ratio.It can accurately extract pitch frequency.