-
Notifications
You must be signed in to change notification settings - Fork 0
/
speech.py
145 lines (137 loc) · 4.8 KB
/
speech.py
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
import numpy as np
import pickle
import hmm
import utils
def normalize(tstack,trstack):
"""
"""
for i in range (len(tstack)):
tstackS = np.column_stack(tstack[i])
mean = np.mean(tstackS,axis = 1)
for j in range(len(tstack[i])):
for k in range ((np.shape(tstack[i][j])[1])):
tstack[i][j][:,k] = tstack[i][j][:,k]-mean
for i in range (len(trstack)):
trstackS = np.column_stack(trstack[i])
mean = np.mean(trstackS,axis = 1)
for j in range(len(trstack[i])):
for k in range ((np.shape(trstack[i][j])[1])):
trstack[i][j][:,k] = trstack[i][j][:,k]-mean
return tstack,trstack
def getlistoflengths(stack):
"""
stack: (b,d,n)
---------------------------------
returns:
lengthlist: n corresponding to each b in stack
"""
tlength = []
Bt = np.shape(stack)[0]
for b in range(Bt):
tlength.append(np.shape(stack[b])[1])
return tlength
# Read in pickled data:
f = open('./data/speech.dat')
data = np.load(f)
# put pickled data in useable form
# print data.keys()
testdata = data.get('test')
traindata = data.get('train')
keyList = testdata.keys()
testdataStack = []
traindataStack = []
for a in keyList:
testdataList = []
traindataList = []
for b in range(len(testdata.get(a))):
testdataList.append((testdata.get(a)[b].T))
for b in range(len(traindata.get(a))):
traindataList.append((traindata.get(a)[b].T))
testdataStack.append((testdataList))
traindataStack.append((traindataList))
testdataStack,traindataStack = normalize(testdataStack, traindataStack)
#print np.shape(testdataStack[0][0])[1]
#llist = getlistoflengths(traindataStack[0])
#print llist
diagcov = False
print "Full Covariance Matrix"
for k in range (1,7):
MarkovModel = []
for a in traindataStack:
trans = hmm.lrtrans(k)
llist = getlistoflengths(a)
MarkovModel.append(hmm.hmm(np.column_stack(a),llist,trans,diagcov = diagcov))
# print np.shape(testdataStack)
# print np.shape(testdataStack[0][0])
TestClassification = []
OriginalTestClassification = []
TrainClassification = []
OriginalTrainClassification = []
for a in range (np.shape(testdataStack)[0]):
llist = getlistoflengths(testdataStack[a])
classifiedtests = []
origtests = []
for b in range (len(llist)):
#origtests.append(keyList[a])
temp = np.array([testdataStack[a][b]])
# print np.shape(temp)
templistlen = []
for c in range (len(keyList)):
templistlen.append(hmm.negloglik(temp,trans = MarkovModel[c][0],dists = MarkovModel[c][1]))
#lassifiedtests.append(keyList[(np.argmin(templistlen))])
TestClassification.append(keyList[(np.argmin(templistlen))])
OriginalTestClassification.append(keyList[a])
llist = getlistoflengths(traindataStack[a])
for b in range (len(llist)):
temp = np.array([traindataStack[a][b]])
templistlen = []
for c in range (len(keyList)):
templistlen.append(hmm.negloglik(temp,trans = MarkovModel[c][0],dists = MarkovModel[c][1]))
TrainClassification.append(keyList[(np.argmin(templistlen))])
OriginalTrainClassification.append(keyList[a])
print "Test Confusion Matrix"
utils.confusion(OriginalTestClassification,TestClassification)
print "Train Confusion Matrix"
utils.confusion(OriginalTrainClassification,TrainClassification)
print k
diagcov = True
print "Diagonal Covariance Matrix"
for k in range (1,7):
MarkovModel = []
for a in traindataStack:
trans = hmm.lrtrans(k)
llist = getlistoflengths(a)
MarkovModel.append(hmm.hmm(np.column_stack(a),llist,trans,diagcov = diagcov))
# print np.shape(testdataStack)
# print np.shape(testdataStack[0][0])
TestClassification = []
OriginalTestClassification = []
TrainClassification = []
OriginalTrainClassification = []
for a in range (np.shape(testdataStack)[0]):
llist = getlistoflengths(testdataStack[a])
classifiedtests = []
origtests = []
for b in range (len(llist)):
#origtests.append(keyList[a])
temp = np.array([testdataStack[a][b]])
# print np.shape(temp)
templistlen = []
for c in range (len(keyList)):
templistlen.append(hmm.negloglik(temp,trans = MarkovModel[c][0],dists = MarkovModel[c][1]))
#lassifiedtests.append(keyList[(np.argmin(templistlen))])
TestClassification.append(keyList[(np.argmin(templistlen))])
OriginalTestClassification.append(keyList[a])
llist = getlistoflengths(traindataStack[a])
for b in range (len(llist)):
temp = np.array([traindataStack[a][b]])
templistlen = []
for c in range (len(keyList)):
templistlen.append(hmm.negloglik(temp,trans = MarkovModel[c][0],dists = MarkovModel[c][1]))
TrainClassification.append(keyList[(np.argmin(templistlen))])
OriginalTrainClassification.append(keyList[a])
print "Test Confusion Matrix"
utils.confusion(OriginalTestClassification,TestClassification)
print "Train Confusion Matrix"
utils.confusion(OriginalTrainClassification,TrainClassification)
print k