forked from danielkrause/DCASE2022-data-generator
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathdb_config.py
189 lines (167 loc) · 11.7 KB
/
db_config.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
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
import numpy as np
import scipy.io
import csv
import librosa
import os
import pickle
class DBConfig(object):
def __init__(
self, params
):
self._rirpath = params['rirpath']
self._mixturepath = params['mixturepath']
self._rirdata = self._load_rirdata()
self._nb_folds = params['nb_folds']
self._rooms2fold = params['rooms2fold']
self._db_path = params['db_path']
self._db_name = params['db_name']
if self._db_name == 'fsd50k':
self._fs = 44100
self._classes = ['femaleSpeech', 'maleSpeech', 'clapping', 'telephone', 'laughter', 'domesticSounds', 'footsteps',
'doorCupboard', 'music', 'musicInstrument', 'waterTap', 'bell', 'knock']
self._nb_classes = len(self._classes)
self._class_mobility = [2, 2, 2, 2, 2, 2, 1, 0, 0, 0, 0, 0, 0]
self._apply_class_gains = True
# self._class_gains = [[0, 0.2004, 0.8008, 6.8766, 357.8846], # femaleSpeech
# [0.0060, 0.4901, 2.5097, 14.3011, 372.2183], # maleSpeech
# [0.3607, 1.1029, 2.6719, 3.9629, 26.6442], #clapping
# [0.0072, 0.8222, 2.3849, 34.1233, 168.5152], #telephone
# [0.0273, 0.8911, 1.9856, 5.6164, 79.1070], #laughter
# [0.0268, 0.1009, 1.8363, 13.9294, 83.2484], #domesticSounds
# [0.0099, 0.3764, 1.2759, 5.4426, 318.8329], #footsteps
# [0.0697, 0.4919, 2.7159, 28.0537, 313.8807], #doorCupboard
# [0.0219, 0.3189, 0.7787, 2.3823, 355.9656], #music
# [0.0160, 0.9563, 2.3413, 5.6720, 168.6679], #musicInstrument
# [0.0972, 0.1828, 0.6304, 0.9522, 125.1975], #waterTap
# [0.0160, 0.9563, 2.3413, 5.6720, 168.6679], #bell
# [0.0697, 0.4919, 2.7159, 28.0537, 313.8807]] #knock
self._class_gains = [[0.0791, 0.5330, 1.3132, 2.2365, 541.3376], # femaleSpeech
[0.0116, 0.6913, 1.2199, 3.0048, 235.0189], # maleSpeech
[0.5083, 2.2579, 3.0934, 7.6387, 100.1174], #clapping
[0.0126, 0.3373, 0.7526, 2.1165, 18.5226], #telephone
[0.1909, 1.4950, 3.2206, 8.2153, 221.2892], #laughter
[0.0004, 1.8347, 3.4778, 5.9276, 555.4895], #domesticSounds
[0.0099, 0.3969, 0.8870, 2.0800, 15.7529], #footsteps
[0.0146, 0.9141, 7.8186, 109.0767, 3979.700], #doorCupboard
[0.1153, 0.4313, 1.2903, 3.3541, 52.6977], #music
[0.0596, 1.4146, 5.3529, 20.6286, 362.0704], #musicInstrument
[0.0117, 0.5505, 1.4926, 2.1936, 44.9466], #waterTap
[0.0596, 1.4146, 5.3529, 20.6286, 362.0704], #bell
[2.4502, 2.4502, 41.3609, 80.2716, 80.2716]] #knock
self._samplelist = self._load_db_fileinfo_fsd()
elif self._db_name == 'nigens':
self._fs = 44100
self._class_dict = {'alarm': 0,'baby': 1, 'crash': 2, 'dog': 3, 'engine': 4, 'femaleScream': 5, 'femaleSpeech': 6,
'fire': 7, 'footsteps': 8, 'knock': 9, 'maleScream': 10, 'maleSpeech': 11,
'phone': 12, 'piano': 13, 'general': 14}
self._class_mobility = [0, 2, 0, 2, 2, 0, 2, 0, 1, 0, 0, 2, 2, 0, 0]
self._classes = list(self._class_dict.keys())
self._nb_classes = len(self._classes)
self._samplelist = self._load_db_fileinfo_nigens()
self._apply_class_gains = False
self._class_gaines = []
def _load_rirdata(self):
with open(os.path.join(self._rirpath, "rirdata_dict.pkl"),'rb') as file:
rirdata = pickle.load(file)
return rirdata
def _load_db_fileinfo_fsd(self):
samplelist_per_fold = []
folds = self._make_selected_filelist()
for nfold in range(self._nb_folds):
print('Preparing sample list for fold {}'.format(str(nfold+1)))
counter = 1
samplelist = {'class': np.array([]), 'audiofile': np.array([]), 'duration': np.array([]), 'onoffset': [], 'nSamples': [],
'nSamplesPerClass': np.array([]), 'meanStdDurationPerClass': np.array([]), 'minMaxDurationPerClass': np.array([])}
for ncl in range(self._nb_classes):
nb_samples_per_class = len(folds[ncl][nfold])
for ns in range(nb_samples_per_class):
samplelist['class'] = np.append(samplelist['class'], ncl)
samplelist['audiofile'] = np.append(samplelist['audiofile'], folds[ncl][nfold][ns])
audiopath = self._db_path + '/' + folds[ncl][nfold][ns]
audio, sr = librosa.load(audiopath)
duration = len(audio)/float(sr)
samplelist['duration'] = np.append(samplelist['duration'], duration)
samplelist['onoffset'].append(np.array([[0., duration],]))
samplelist['nSamples'].append(counter)
counter += 1
samplelist['onoffset'] = np.squeeze(np.array(samplelist['onoffset'],dtype=object))
for n_class in range(self._nb_classes):
class_idx = (samplelist['class'] == n_class)
samplelist['nSamplesPerClass'] = np.append(samplelist['nSamplesPerClass'], np.sum(class_idx))
if n_class == 0:
samplelist['meanStdDurationPerClass'] = np.array([[np.mean(samplelist['duration'][class_idx]), np.std(samplelist['duration'][class_idx])]])
samplelist['minMaxDurationPerClass'] = np.array([[np.min(samplelist['duration'][class_idx]), np.max(samplelist['duration'][class_idx])]])
else:
samplelist['meanStdDurationPerClass'] = np.vstack((samplelist['meanStdDurationPerClass'], np.array([np.mean(samplelist['duration'][class_idx]), np.std(samplelist['duration'][class_idx])])))
samplelist['minMaxDurationPerClass'] = np.vstack((samplelist['minMaxDurationPerClass'], np.array([np.min(samplelist['duration'][class_idx]), np.max(samplelist['duration'][class_idx])])))
samplelist_per_fold.append(samplelist)
return samplelist_per_fold
def _load_db_fileinfo_nigens(self):
samplelist_per_fold = []
for nfold in range(self._nb_folds):
print('Preparing sample list for fold {}'.format(str(nfold+1)))
foldlist_file = self._db_path + '/NIGENS_8-foldSplit_fold' + str(nfold+1) + '_wo_timit.flist'
filelist = []
with open(foldlist_file, newline = '') as flist:
flist_reader = csv.reader(flist, delimiter='\t')
for fline in flist_reader:
filelist.append(fline)
flist_len = len(filelist)
samplelist = {'class': np.array([]), 'audiofile': np.array([]), 'duration': np.array([]), 'onoffset': [], 'nSamples': flist_len,
'nSamplesPerClass': np.array([]), 'meanStdDurationPerClass': np.array([]), 'minMaxDurationPerClass': np.array([])}
for file in range(flist_len):
clsfilename = filelist[file][0].split('/')
clsname = clsfilename[0]
filename = clsfilename[1]
samplelist['class'] = np.append(samplelist['class'], int(self._class_dict[clsname]))
samplelist['audiofile'] = np.append(samplelist['audiofile'], clsname + '/' + filename)
audiopath = self._db_path + '/' + clsname + '/' + filename
#print(audiopath)
#with contextlib.closing(wave.open(audiopath,'r')) as f:
audio, sr = librosa.load(audiopath)
samplelist['duration'] = np.append(samplelist['duration'], len(audio)/float(sr))
if clsname == 'general':
onoffsets = []
onoffsets.append([0., samplelist['duration'][file]])
samplelist['onoffset'].append(np.array(onoffsets))
else:
meta_file = self._db_path + '/' + clsname + '/' + filename + '.txt'
onoffsets = []
with open(meta_file, newline = '') as meta:
meta_reader = csv.reader(meta, delimiter='\t')
for onoff in meta_reader:
onoffsets.append([float(onoff[0]), float(onoff[1])])
samplelist['onoffset'].append(np.array(onoffsets))
samplelist['onoffset'] = np.squeeze(np.array(samplelist['onoffset'],dtype=object))
for n_class in range(self._nb_classes):
class_idx = (samplelist['class'] == n_class)
samplelist['nSamplesPerClass'] = np.append(samplelist['nSamplesPerClass'], np.sum(class_idx))
if n_class == 0:
samplelist['meanStdDurationPerClass'] = np.array([[np.mean(samplelist['duration'][class_idx]), np.std(samplelist['duration'][class_idx])]])
samplelist['minMaxDurationPerClass'] = np.array([[np.min(samplelist['duration'][class_idx]), np.max(samplelist['duration'][class_idx])]])
else:
samplelist['meanStdDurationPerClass'] = np.vstack((samplelist['meanStdDurationPerClass'], np.array([np.mean(samplelist['duration'][class_idx]), np.std(samplelist['duration'][class_idx])])))
samplelist['minMaxDurationPerClass'] = np.vstack((samplelist['minMaxDurationPerClass'], np.array([np.min(samplelist['duration'][class_idx]), np.max(samplelist['duration'][class_idx])])))
samplelist_per_fold.append(samplelist)
return samplelist_per_fold
def _make_selected_filelist(self):
folds = []
folds_names = ['train', 'test'] #TODO: make it more generic
nb_folds = len(folds_names)
class_list = self._classes #list(self._classes.keys())
for ntc in range(self._nb_classes):
classpath = self._db_path + '/' + class_list[ntc]
per_fold = []
for nf in range(nb_folds):
foldpath = classpath + '/' + folds_names[nf]
foldcont = os.listdir(foldpath)
nb_subdirs = len(foldcont)
filelist = []
for ns in range(nb_subdirs):
subfoldcont = os.listdir(foldpath + '/' + foldcont[ns])
for nfl in range(len(subfoldcont)):
if subfoldcont[nfl][0] != '.' and subfoldcont[nfl].endswith('.wav'):
filelist.append(class_list[ntc] + '/' + folds_names[nf] + '/' + foldcont[ns] + '/' + subfoldcont[nfl])
per_fold.append(filelist)
folds.append(per_fold)
return folds