hello
this is my code but there are a few open questions
- why you say that we take 5 samples while y and z are vectors of 4 samples
- I suppose it's a typo but I guess y3= [x3 x2 x4 x5] is wrong and should be y3= [x3 x4 x5 x6]
- the resulting vector is about 4 times longer than the input vector but that's normal as each x vaues get's replicated multiple times in the output (z)
clc
clearvars
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
% load signal
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% data
[signal,Fs] = audioread('test_voice_mono.wav');
[samples,channels] = size(signal);
dt = 1/Fs;
time = (0:samples-1)*dt;
% extract four samples buffers with one sample shift
shift = 1; % nb of samples shift between successive buffers
buffer = 4; % nb of samples in one buffer
N = fix((samples-buffer)/shift +1);
z_all = [];
for ci=1:N
start_index = 1+(ci-1)*shift;
stop_index = min(start_index+ buffer-1,samples);
y =signal(start_index:stop_index,:); % 4 contiguous samples (y1 = [x1 x2 x3 x4])
% flip values so that it becomes : z1 = [x3 x2 x4 x1]
ind = [3;2;4;1];
z = y(ind);
% concatenate all z vectors
z_all = [z_all; z];
end
% audiowrite('out.wav',z_all,Fs);
sound(z_all,Fs)
