s3dman / aex Goto Github PK

1- Collect N_Samples Vector from sound card 2- Appli window function to N_Samples Vector (Hann aka Hanning window) 3- Set a ComplexVector ( Real = N_Samples , Img = 0) 4- Appli FFT(ComplexVector)

5- Set peak frequencies in the spectrum array for(j = 0; j ＜(N/2) ; j++){ magnitude = sqrt( ComplexVector[j].re^2+ ComplexVector[j].img^2); freq = j * SampleRate /N; for( i=0;i＜BUCKETS;i++){ if( freq＞=Freq_Bin[i] && freq＜=Freq_Bin[i+1] ){ if (magnitude ＞ Spectrum[i]){ Spectrum[i] = magnitude; } } } } Freq_Bin: is a distribution of audible frequencies

6- If you want appli Spectrum deformations f(x)

Spectrum[i]= f( Spectrum[i] )

Natural f(x)=x*Fit_factor

Exponential f(x)=log(x*Fit_factor2)*Fit_factor

Multi Peak Scale f(x,i)=x/Peak[i]*Fit_factor

Max Peak Scale f(x)=x/Global_Peak*Fit_factor

7- Render spectrum magnitudes (Spectrum[i])