MFCC by librosa

import librosa

import librosa.display

# Generate mfccs from a time series

y, sr = librosa.load('0acdb43dae5a7a66945bc44ee87ee67cd64f4e9dabf2817ab59ffe8bf9093778.wav')

librosa.feature.mfcc(y=y, sr=sr)

# array([[ -5.229e+02,  -4.944e+02, ...,  -5.229e+02,  -5.229e+02],

# [  7.105e-15,   3.787e+01, ...,  -7.105e-15,  -7.105e-15],

# ...,

# [  1.066e-14,  -7.500e+00, ...,   1.421e-14,   1.421e-14],

# [  3.109e-14,  -5.058e+00, ...,   2.931e-14,   2.931e-14]])

# Use a pre-computed log-power Mel spectrogram

S = librosa.feature.melspectrogram(y=y, sr=sr, n_mels=128,

                                   fmax=8000)

librosa.feature.mfcc(S=librosa.power_to_db(S))

# array([[ -5.207e+02,  -4.898e+02, ...,  -5.207e+02,  -5.207e+02],

# [ -2.576e-14,   4.054e+01, ...,  -3.997e-14,  -3.997e-14],

# ...,

# [  7.105e-15,  -3.534e+00, ...,   0.000e+00,   0.000e+00],

# [  3.020e-14,  -2.613e+00, ...,   3.553e-14,   3.553e-14]])

# Get more components

mfccs = librosa.feature.mfcc(y=y, sr=sr, n_mfcc=40)

# Visualize the MFCC series

import matplotlib.pyplot as plt

fig = plt.figure(figsize=(4, 4))

librosa.display.specshow(mfccs)

#plt.colorbar()

#plt.title('MFCC')

plt.tight_layout()

fig.savefig('test.png')

plt.show()

# save png using pylab

#fig= plt.figure()