tabber

% Tabber
% Elijah Rockers - 2013
% Autocorrelation
 
% E2 = 82.41Hz
% E6 = 1318.510Hz
 
x = wavread('G Major.wav');
 
% Variables
sample_rate = 48000;
window_size = 2000;
min_lag = 36;
max_lag = 582;
 
% Roughly calculates how many iterations are needed to cover all data
j = (size(x, 1) / (max_lag));
j = floor(j);
 
for i = 1 : j
% Makes sure 'x1' does not exceed valid size
if ((size(x, 1)) < ((max_lag * i) + window_size - max_lag))
break
 
else
x1 = x(((max_lag * i) + 1 - max_lag) : ((max_lag * i) + window_size - max_lag));
% May need to 'fade window' x1 here
% This for - loop autocorrelates
for k = min_lag : max_lag
% Make sure 'x2' doesn't exceed valid size
if ((size(x, 1)) < ((max_lag * i) + window_size - max_lag + k))
break
else
x2 = x(((max_lag * i) + 1 - max_lag + k) : ((max_lag * i) + window_size - max_lag + k));
r(i, k) = dot(x1, x2);
end
end
% 'n' keeps track of which sample each window started on
n(i) = (max_lag * i) + 1 - max_lag;
max_i(i) = 0;
% THIS NEEDS TO FIND PERIOD... DOESN'T REALLY
for k = 2 : (size(r, 2) - 1)
if (size(r, 1) < i)
break
else
if ((r(i, k) > r(i, k - 1)) && (r(i, k) > r(i, k + 1)))
max_i(i) = k;
end
end
end
freq_i(i) = (1 / (max_i(i) / sample_rate));
end
end
 
for i = 1 : size(freq_i, 2)
if freq_i(i) > 1320
freq_i(i) = 0;
end
end
 
figure(1);
subplot(2, 1, 1), plot(1 : size(x), x(1 : size(x)));
xlabel('samples');
ylabel('audio');
subplot(2, 1, 2), plot(n, freq_i(1 : size(freq_i, 2)));
xlabel('samples');
ylabel('frequency');