Digital-Audio-Processing

Scripts to process digital signal(Audio).

features.py

This script extracts features from audio. it extracts time-domain, frequency-domain, and frequency-time-domain features. it is inspired by Valerio Velardo's playlist on YouTube Valerio Velardo list, and his code on githubValerio Velardo code

time-domain features:

• Amplitude Envelope Feature(AEt):
• Root-Mean-Square Energy Feature(RMSt):
• Zero Crossing Rate Feature(ZCRt):

Frequency-domain features:

Fast Fourier Transform (FFT):

• Band Energy Ratio:
• Spectral Centroid:
• Bandwidth (Spectral spread):

Frequency-time Domin Features:

• Spectrogram:
• Mel Spectrogram:
• Mel-Frequency Cepstral Coefficients(MFCCs):
• Delta of MFCCs:
• Delta 2 of MFCCs: