-
Notifications
You must be signed in to change notification settings - Fork 0
/
learn_svdtmap.m
193 lines (174 loc) · 9.15 KB
/
learn_svdtmap.m
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
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
clear all;
% Create the mapping table first by running
load FileWav16kHz
% see create_gmm for what to be loaded by FileWav16kHz
doNormalize = 1;
SaveName = sprintf('SvdClassTm');
if doNormalize
SaveName = [SaveName, 'N'];
end
SampleOffset = 120; % Discard some samples (at the end, most samples are impulsive)
% Load IsTrain and IsTest from another file
load SpkrSubset
%% Class definition
% Gender
Class(1, 1).name = 'Male';
Class(1, 1).path = '.\MappingTable\GT100\N3W180R0S0EdN1\english]english239.male.N_english.R_usa.Y18.A19';
Class(1, 1).IsTrain = remove_samepath(IsTrain.IsMale, Class(1, 1).path, accnames);
Class(1, 1).IsTest = remove_samepath(IsTest.IsMale, Class(1, 1).path, accnames);
Class(1, 2).name = 'Female';
Class(1, 2).path = '.\MappingTable\GT100\N3W180R0S0EdN1\english]english10.female.N_english.R_usa.Y35.A35';
Class(1, 2).IsTrain = remove_samepath(IsTrain.IsFemale, Class(1, 2).path, accnames);
Class(1, 2).IsTest = remove_samepath(IsTest.IsFemale, Class(1, 2).path, accnames);
% Native Male
Class(2, 1).name = 'NativeMaleRef';
Class(2, 1).path = '.\MappingTable\GT100\N3W180R0S0EdN1\english]english451.male.N_english.R_usa.Y44.A44';
Class(2, 1).IsTrain = remove_samepath(IsTrain.IsNative, Class(2, 1).path, accnames);
Class(2, 1).IsTest = remove_samepath(IsTest.IsNative, Class(2, 1).path, accnames);
Class(2, 2).name = 'NonNativeMaleRef';
Class(2, 2).path = '.\MappingTable\GT100\N3W180R0S0EdN1\japanese]japanese4.male.N_japanese.R_usa.Y1.A20';
Class(2, 2).IsTrain = remove_samepath(IsTrain.IsNonNative, Class(2, 2).path, accnames);
Class(2, 2).IsTest = remove_samepath(IsTest.IsNonNative, Class(2, 2).path, accnames);
%
% % Native Female
Class(3, 1).name = 'NativeFemaleRef';
Class(3, 1).path = '.\MappingTable\GT100\N3W180R0S0EdN1\english]english165.female.N_english.R_usa.Y43.A43';
Class(3, 1).IsTrain = remove_samepath(IsTrain.IsNative, Class(3, 1).path, accnames);
Class(3, 1).IsTest = remove_samepath(IsTest.IsNative, Class(3, 1).path, accnames);
Class(3, 2).name = 'NonNativeFemaleRef';
Class(3, 2).path = '.\MappingTable\GT100\N3W180R0S0EdN1\japanese]japanese26.female.N_japanese.R_usa.Y6.A44';
Class(3, 2).path = '.\MappingTable\GT100\N3W180R0S0EdN1\japanese]japanese13.male.N_japanese.R_usa.Y0.A28';
Class(3, 2).IsTrain = remove_samepath(IsTrain.IsNonNative, Class(3, 2).path, accnames);
Class(3, 2).IsTest = remove_samepath(IsTest.IsNonNative, Class(3, 2).path, accnames);
%% Modified Class Definition
% Class(3, 1).IsTrain = Class(3, 1).IsTrain.*SpkrSubset(1, 3).bool;
% Class(3, 1).IsTest = Class(3, 1).IsTest.*SpkrSubset(1, 3).bool;
% Class(3, 2).IsTrain = Class(3, 2).IsTrain.*SpkrSubset(2, 3).bool;
% Class(3, 2).IsTest = Class(3, 2).IsTest.*SpkrSubset(2, 3).bool;
%% Create Concatenated vector
% GTdiff2, GTpoly1, choose 1 !
% Polynomial MicrotimingDifference PhnDifference WrdDifference, Choose one
% of normalizing kind
AV = accdata_read(Class, accnames, 'Train', SampleOffset);
AVtest = accdata_read(Class, accnames, 'Test', SampleOffset);
%% Training SVD
[NClassDef, NGroup] = size(Class);
for iClassDef = 1:NClassDef
for FromGroup = 1:NGroup
for RefGroup = 1 : NGroup
X1 = AV(iClassDef, FromGroup, RefGroup).VecsDurMicro;
X2 = AV(iClassDef, FromGroup, RefGroup).VecsDurPhn;
X3 = AV(iClassDef, FromGroup, RefGroup).VecsDurWrd;
X4 = AV(iClassDef, FromGroup, RefGroup).VecsDurMicroNosp;
X5 = AV(iClassDef, FromGroup, RefGroup).VecsDurPhnNosp;
X6 = AV(iClassDef, FromGroup, RefGroup).VecsDurWrdNosp;
[U1, S, V] = svds(X1, 100);
[U2, S, V] = svds(X2, 100);
[U3, S, V] = svds(X3, 100);
[U4, S, V] = svds(X4, 100);
[U5, S, V] = svds(X5, 100);
[U6, S, V] = svds(X6, 100);
SvdClass(iClassDef, FromGroup, RefGroup).U1 = U1; % Too large to save
SvdClass(iClassDef, FromGroup, RefGroup).U2 = U2; % Too large to save
SvdClass(iClassDef, FromGroup, RefGroup).U3 = U3; % Too large to save
SvdClass(iClassDef, FromGroup, RefGroup).U4 = U4; % Too large to save
SvdClass(iClassDef, FromGroup, RefGroup).U5 = U5; % Too large to save
SvdClass(iClassDef, FromGroup, RefGroup).U6 = U6; % Too large to save
end
end
end
% norm(SvdClass(1, 1, 1).U - SvdClass(1, 2, 1).U)
% norm(SvdClass(1, 1, 2).U - SvdClass(1, 2, 2).U)
% norm(SvdClass(1, 1, 1).U - SvdClass(1, 2, 1).U)
% fprintf('SVD Computation completed and Saving the data\n');
% save(SaveName, 'SvdClass', 'Class', 'SampleOffset');
%% SVD Classification Rule
% For Gender, There are 4 different groups of features,
% A1) M -> M; 1 -> 1
% A2) F -> M; 2 -> 1
% B1) M -> F; 1 -> 2
% B2) F -> F; 2 -> 1
% For Male Speaker,
% test Err(X | M -> M) < Err(X | F -> M)
% +) test Err(X | M -> F) < Err(X | F -> F)
% For Female Speaker
% test Err(X | M -> M) > Err(X | F -> M)
% +) test Err(X | M -> F) > Err(X | F -> F)
% Generalize
% test Err(X1 | A1 ) < Err(X1 | A2 )
% +) test Err(X1 | B1 ) < Err(X1 | B2 )
%
% test Err(X2 | A1 ) > Err(X2 | A2 )
% +) test Err(X2 | B1 ) > Err(X2 | B2 )
%
% load SvdClassTmN
NSvd =60;
[NClassDef, NGroup] = size(Class);
clear Err;
for iClassDef = 1 : NClassDef
for TestGroup = 1 : NGroup
for RefGroup = 1 : NGroup
for FromGroup = 1 : NGroup
Fidx = find(Class(iClassDef, TestGroup).IsTest);
U = SvdClass(iClassDef, FromGroup, RefGroup).U1(:, 1:NSvd);
% U = SvdClass(iClassDef, FromGroup, RefGroup).U2(:, 1:NSvd);
% U = SvdClass(iClassDef, FromGroup, RefGroup).U3(:, 1:NSvd);
% U = SvdClass(iClassDef, FromGroup, RefGroup).U4(:, 1:NSvd);
% U = SvdClass(iClassDef, FromGroup, RefGroup).U5(:, 1:NSvd);
% U = SvdClass(iClassDef, FromGroup, RefGroup).U6(:, 1:NSvd);
U_ = inv(U'*U)*U';
for i = 1:length(Fidx)
Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurMicro(:, i);
% Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurPhn(:, i);
% Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurWrd(:, i);
% Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurMicroNosp(:, i);
% Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurPhnNosp(:, i);
% Vec = AVtest(iClassDef, TestGroup, RefGroup).VecsDurWrdNosp(:, i);
% fname = fullfile(Class(iClassDef, RefGroup).path, [accnames{Fidx(i)}, '.mat']);
% load(fname); % load 'GTpoly1'
% Coef = U_*GTpoly1(SampleOffset:end-SampleOffset);
Coef = U_*Vec;
% if 0
% [a, b, c] = fileparts(Class(iClassDef, RefGroup).path);
% fnameRef = fullfile('wav16kHzMlfDur', [b, c, '.mat']);
% V = load(fnameRef);
% % GTnew = mt2gt(MT, V.t1*10e-8*16000); % Phn
% GTnew = mt2gt(MT, V.wt1*10e-8*16000); % Wrd
% Vec = GTnew(:, 1);
% Vec = Vec - Vec(2);
% Vec = Vec/Vec(end);
% Vec = diff(Vec);
% Coef = U_*Vec;
% else % Pure mapping
% Coef = U_*GTdiff2(SampleOffset:end-SampleOffset, 1);
% end
VecHat = U*Coef;
% Err(iClassDef, TestGroup, FromGroup, RefGroup, i).n2...
% = norm(VecHat - GTpoly1(SampleOffset:end-SampleOffset), 2)/sqrt(length(VecHat)); % Same for any other distance : Err(TestGroup, FromGroup, RefGroup, j).n2 = simmx2(VecHat, GTpoly1, 'corrcoef');
% Err(iClassDef, TestGroup, FromGroup, RefGroup, i).n2...
% = norm(VecHat - GTdiff2(SampleOffset:end-SampleOffset, 1), 2)/sqrt(length(VecHat)); % Same for any other distance : Err(TestGroup, FromGroup, RefGroup, j).
Err(iClassDef, TestGroup, FromGroup, RefGroup, i).n2...
= norm(VecHat - Vec, 2)/sqrt(length(VecHat)); % Same for any other distance : Err(TestGroup, FromGroup, RefGroup, j).
end
end
end
end
end
for iClassDef = 1 : NClassDef
for TestGroup = 1 : NGroup
A1 = [Err(iClassDef, TestGroup, 1, 1, :).n2];
A2 = [Err(iClassDef, TestGroup, 2, 1, :).n2];
B1 = [Err(iClassDef, TestGroup, 1, 2, :).n2];
B2 = [Err(iClassDef, TestGroup, 2, 2, :).n2];
[MinVal, MinIdx] = min([A1; A2; B1; B2]);
% Report the percent correct if MinIdx is either A1 or A2, i.e., 1 or 3
if TestGroup == 1
PC(iClassDef, TestGroup) = (sum(MinIdx==1) + sum(MinIdx==3))/length(MinIdx);
elseif TestGroup == 2
PC(iClassDef, TestGroup) = (sum(MinIdx==2) + sum(MinIdx==4))/length(MinIdx);
end
end
end
PC
K = mean(PC(2:3, :));
K*[1/4;3/4]