Pitch-Estimation-and-Voicing-Detection/dataClassificationAnalysis.m at master · Gonaco/Pitch-Estimation-and-Voicing-Detection · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
clc
close all
clear all

files = dir('data/fda_ue/*.f0ref');
audios = dir('data/fda_ue/*.wav');

n_files = length(files);

w_prior = [];
z_array = [];

figure

for f=1:n_files

    w_prior = [w_prior; textread(strcat('data/fda_ue/',files(f).name))];

    [x, f_s] = audioread(strcat('data/fda_ue/',audios(f).name));

    x = expandVector(x,15,f_s); % Fill the signal x with 0 15ms in the begining and 15ms in the end

    sampl_numb = length(x);
    audio_t = (sampl_numb/f_s)*1e3; %in ms
    w_L = 32; % 32ms window

    %max_array = []
    for counter = 0:15:(audio_t-w_L)

        start = f_s*counter/1e3+1;
        finish = f_s*(counter+w_L)/1e3+1;

        x_w = x(start:finish);

        zeroes_n = 0;

        for k = 2:(w_L*f_s*1e-3)
            if x_w(k) == 0
                zeroes_n = zeroes_n+1;
            elseif (x_w(k) > 0 && x_w(k-1) < 0)
                zeroes_n = zeroes_n+1;
            elseif (x_w(k) < 0 && x_w(k-1) > 0)
                zeroes_n = zeroes_n+1;
            end
        end

        z_array = [z_array; zeroes_n];

    end

    if length(w_prior) - length(z_array) == 1
        w_prior = w_prior(1:end-1);
    elseif length(w_prior) - length(z_array) == -1
        z_array = z_array(1:end-1);
    end

end

voiced = w_prior;
voiced(voiced>0) = 1;

labelsVoiced = find(voiced>0);
voicedSamples = z_array(labelsVoiced);
labelsUnvoiced = find(voiced<1);
unvoicedSamples = z_array(labelsUnvoiced);

subplot(3,1,1)
hist(voicedSamples, 500)
hold on
[a b] = hist(unvoicedSamples, 500);
ab=bar(b,a);
set(ab,'FaceColor','none','EdgeColor','r');
hold off
title('Histogram. Number of zeroes')

x=1:600;
likelihood_voiced = 1/(sqrt(2*pi*var(voicedSamples)))*exp(-1/(2*var(voicedSamples))*(x-mean(voicedSamples)).^2);
likelihood_unvoiced = 1/(sqrt(2*pi*var(unvoicedSamples)))*exp(-1/(2*var(unvoicedSamples))*(x-mean(unvoicedSamples)).^2);

[mu, sigma] = normfit(voicedSamples);
likelihood_voiced = 1/(sigma*sqrt(2*pi))*exp(-1/2*((x-mu)/sigma).^2);
likelihood_voiced = normpdf(x,mu,sigma/2);

[mu, sigma] = normfit(unvoicedSamples);
likelihood_unvoiced = normpdf(x,mu,sigma/2);

prior_voiced = length(labelsVoiced)/(length(labelsVoiced)+length(labelsUnvoiced));
prior_unvoiced = length(labelsUnvoiced)/(length(labelsVoiced)+length(labelsUnvoiced));

evidence = 1/length(x);

subplot(3,2,3)
histfit(voicedSamples, 500)
title('Voiced histogram approximation')

subplot(3,2,4)
histfit(unvoicedSamples, 500)
title('Unvoiced histogram approximation')

subplot(3,2,5)
plot(likelihood_voiced)
hold on
plot(likelihood_unvoiced, 'r')
hold off
title('p(x|w_i)')

voiced_gauss = likelihood_voiced.*prior_voiced;
unvoiced_gauss = likelihood_unvoiced.*prior_unvoiced;
subplot(3,2,6)
plot(voiced_gauss)
hold on
plot(unvoiced_gauss, 'r')
hold off
title('p(w_i|x)')

P_e = sum(unvoiced_gauss(1:99)) + sum(voiced_gauss(99:end))