How to generate an audio file from the MFCC features

Question

I have extracted MFCC features using python_speech_features

The idea is to generate an audio file from the MFCC features.

import librosa
import python_