在多數(shù)的現(xiàn)代語音識別系統(tǒng)中，人們都會用到頻域特征。梅爾頻率倒譜系數(shù)（MFCC）,首先計算信號的功率譜，然后用濾波器和離散余弦變換的變換來提取特征。本文重點介紹如何提取MFCC特征。

首先創(chuàng)建有一個Python文件，并導入庫文件：? ? ?from scipy.io import wavfile? ? ?from python_speech_features import mfcc, logfbank? ? ?import matplotlib.pylab as plt1、首先創(chuàng)建有一個Python文件，并導入庫文件：? ? ?from scipy.io import wavfile? ? ?from python_speech_features import mfcc, logfbank? ? ?import matplotlib.pylab as plt

讀取音頻文件：

samplimg_freq, audio = wavfile.read("data/input_freq.wav")

提取MFCC特征和過濾器特征：

?????mfcc_features = mfcc(audio, samplimg_freq)

? ? ?filterbank_features = logfbank(audio, samplimg_freq)

打印參數(shù)，查看可生成多少個窗體：

            
   print('\nMFCC:\nNumber of windows =', mfcc_features.shape[0])

   print('Length of each feature =', mfcc_features.shape[1])

   print('\nFilter bank:\nNumber of windows=', filterbank_features.shape                                                         [0])

   print('Length of each feature =', filterbank_features.shape[1])

          

將MFCC特征可視化。轉換矩陣，使得時域是水平的：

            
   mfcc_features = mfcc_features.T

   plt.matshow(mfcc_features)

   plt.title('MFCC')

          

將濾波器組特征可視化。轉化矩陣，使得時域是水平的：

            
   filterbank_features = filterbank_features.T

   plt.matshow(filterbank_features)

   plt.title('Filter bank')

   

   plt.show()