Working with Audio in Python

 What I did is read a .wav file, extract the raw audio data, (I use first 1/20 data for sample), transform it using Fast Fourier Transform, manipulate it (I only use the first cluster frequency spectrum ), transform it back using InverseFFT.

 I use matplotlib and numpy module to plot and compute.

 I also use sys and wave module as 'interface'.

import matplotlib.pyplot as plt
import numpy as np
import wave
import sys


spf = wave.open('violin2.wav','r')

#Extract Raw Audio from Wav File
signal = spf.readframes(-1)
signal = np.fromstring(signal, 'Int16')
fs = spf.getframerate()
print fs

#If Stereo
if spf.getnchannels() == 2:
    print 'Just mono files'
    sys.exit(0)

Time=np.linspace(0, len(signal)/fs, num=len(signal))
sample = []
st=[]
for i in np.arange(len(signal)/20):
    sample.append( signal[i])
    st.append(Time[i])

ft=np.fft.fft(sample)
pft=[]
for i in np.arange(len(ft)):
    if i<500: data-blogger-escaped-ample="" data-blogger-escaped-else:="" data-blogger-escaped-from="" data-blogger-escaped-ft="" data-blogger-escaped-i="" data-blogger-escaped-ift="" data-blogger-escaped-ignal="" data-blogger-escaped-ime="" data-blogger-escaped-pft.append="" data-blogger-escaped-pft="" data-blogger-escaped-plt.figure="" data-blogger-escaped-plt.grid="" data-blogger-escaped-plt.plot="" data-blogger-escaped-plt.show="" data-blogger-escaped-plt.title="" data-blogger-escaped-pre="" data-blogger-escaped-rue="" data-blogger-escaped-sample="" data-blogger-escaped-signal="" data-blogger-escaped-st="" data-blogger-escaped-wave...="">

.