main.m

function main(ip, fsr, g_max, p_sat, tr_1, tr_2, tr_3, tr_4)

% Main File

%ip  : input
%fsr : frequency sampling rate
%g_max : maximum gain to be applied
%p_sat : Saturation power
%tr_ : Frequency transition values

%ip = ip(:,1); % Taking one channel of the audio

figure;
plot(ip),title('Input Signal')

%Applying a low-pass filter with frequency of 20,000Hz which is maximum
%frequency audible to normal human ears.

f_pass = 20000;
ip2 = lowpass(ip,f_pass,fsr);


% Addition of white gaussian noise to the input signal
%
%Additive white Gaussian noise is a basic noise model used in 
%Information theory to mimic the effect of many random processes that occur in nature. 
%
%White Noise: A random process signal with a constant 
%power spectral density (PSD) function is a white noise process.

% awgn(in,snr) : snr : signal to noise ratio
ip3 = awgn(ip2,30);
figure;
plot(ip3),title('Input Signal after addition of Noise')


% denoising filter
ip_de_n = Denoise(ip3); 


% frequency shaping filter

ip_f = FreqShape(ip_de_n,g_max,tr_1,tr_2,tr_3,tr_4,fsr);

% amplitude compression filter

ip_a = AmplitudeFilter(ip_f,p_sat,fsr);     
figure;
plot(ip_a),title('Input Signal after amplification')

% analysis
ip_length = length(ip);
t=0:1/fsr:(ip_length-1)/fsr;

figure;
subplot(2,1,1);
plot(t,ip,'b');
axis tight;
xlabel('Time (sec)');
ylabel('Relative Magnitude');
title('Time Profile for Data in Signal 2');


subplot(2,1,2);
plot(t,ip_a,'r');
axis tight;
xlabel('Time (sec)');
ylabel('Relative Magnitude');
title('Time Profile for Data in Adjusted Signal 2');

figure;
subplot(2,1,1);
specgram(ip);
title('Spectrogram of Original Signal 2');

subplot(2,1,2);
specgram(ip_a);
title('Spectrogram of Adjusted Signal 2');



soundsc(ip_a,fsr);
%auwrite(ip_a,fsr,info.BitsPerSample,'linear',newfile);