i too got the same problem while performing speaker recognition using vq method.......here is the code and the error is ??? Subscript indices must either be real positive integers or logicals.
Error in ==> vqlbg at 30 c(:, i) = mean(d(:, (find(ind == j))), 2); code: clc clear all; close all; code=train('C:\Users\welcome\Desktop\project\matlab\train\',8); test('C:\Users\welcome\Desktop\project\matlab\test\',8,code); train.m: function code = train(traindir, n)
k = 16;
for i = 1:n
file = sprintf('%ss%d.wav', traindir, i);
disp(file);
[s, fs] = wavread(file);
size(s)
v = melfcc(s, fs);
code{i} = vqlbg(v, k);
end
test.m
function test(testdir, n, code)
for l = 1:length(code)
d = disteu(v, code{l});
dist = sum(min(d,[],2))/size(d,l);
if dist1< distmin
distmin = dist;
k1 = l;
end
end
msg = sprintf('Speaker %d matches with speaker %d', k, k1);
disp(msg);
melfcc.m:
function r = melfcc(s, fs1)
m=160;
n=256;
frames=blockFrames(s,fs1,m,n);
t=n/2;
tmax=length(s)/fs1;
m=melfb(n,16,fs1)
n2=1+floor(n/2)
length(m);
length(frames)
disp(frames)
r=dct(log(abs(frames)));
end
blockFrames.m:
function M3 = blockFrames(s, fs, m, n)
l = length(s);
nbFrame = floor((l - n)/m) + 1;
for i = 1:n
for j = 1:nbFrame
M(i, j) = s(((j - 1) * m) + i);
end
end
h = hamming(n);
M2 = diag(h) * M;
for i = 1:nbFrame
M3(:, i) = fft(M2(:, i));
end
end
size(code{1})
melfb.m:
function m = melfb(p, n, fs)
f0 = 700 / fs;
fn2 = floor(n/2);
lr = log(1 + 0.5/f0) / (p+1);
bl = n * (f0 * (exp([0 1 p p+1] * lr) - 1));
b1 = floor(bl(1)) + 1;
b2 = ceil(bl(2));
b3 = floor(bl(3));
b4 = min(fn2, ceil(bl(4))) - 1;
pf = log(1 + (b1:b4)/n/f0) / lr;
fp = floor(pf);
pm = pf - fp;
r = [fp(b2:b4) 1+fp(1:b3)];
c = [b2:b4 1:b3] + 1;
v = 2 * [1-pm(b2:b4) pm(1:b3)];
m = sparse(r, c, v, p, 1+fn2);
end
disteu.m:
function d = disteu(x, y)
[M, N] = size(x);
[M2, P] = size(y);
if (M ~= M2) error('Matrix dimensions do not match.') end
d = zeros(N, P);
if (N < P) copies = zeros(1,P); for n = 1:N d(n,:) = sum((x(:, n+copies) - y) .^2, 1); end else copies = zeros(1,N); for p = 1:P d(:,p) = sum((x - y(:, p+copies)) .^2, 1)'; end end
d = d.^0.5; end vqlbg.m: function c =vqlbg(d, k) % VQLBG Vector quantization using the Linde-Buzo-Gray algorithm % % Inputs: % d contains training data vectors (one per column) % k is number of centroids required % % Outputs: % c contains the result VQ codebook (k columns, one for each % centroids)
%Hints
%Cluster Vectors (Nearest-Neighbor Search): c=mean(d,2);
%The nearest-neighbor search step is: given a current codebook c, assign each training vector in d with a closest codeword. To do that, one needs to compute the pair-wise distances between each vectors in d to each vectors (codeword) in c. This can be done with the supplied function disteu: z = disteu(d, c);
%Now z(i, j) would be the distance between the training vector d(:, i) and the codeword c(:, j). Next step, for each training vector, find the closest codeword. For this, use the Matlab function min:
[m, ind] = min(z, [], 2);
d(:, find(ind == j));
%Find Centroids
%The centroid of all vectors in a particular cluster is found by the Matlab function mean. For example, after the Nearest-Neighbor Search step above, the new centroids of the clusters number j can be updated as follows: c(:, i) = mean(d(:, (find(ind == j))), 2); % c(:, j) = mean(d(:, b, 2)); end plssssssssssssshelp urgent
