-
Notifications
You must be signed in to change notification settings - Fork 0
/
data_generator.py
338 lines (306 loc) · 13.3 KB
/
data_generator.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
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
"""
Defines a class that is used to featurize audio clips, and provide
them to the network for training or testing.
"""
import json
import numpy as np
import random
from python_speech_features import mfcc
import librosa
import scipy.io.wavfile as wav
import matplotlib.pyplot as plt
from mpl_toolkits.axes_grid1 import make_axes_locatable
from utils import calc_feat_dim, spectrogram_from_file, text_to_int_sequence
from utils import conv_output_length
RNG_SEED = 123
class AudioGenerator():
def __init__(self, step=10, window=20, max_freq=8000, mfcc_dim=13,
minibatch_size=20, desc_file=None, spectrogram=True, max_duration=10.0,
sort_by_duration=False):
"""
Params:
step (int): Step size in milliseconds between windows (for spectrogram ONLY)
window (int): FFT window size in milliseconds (for spectrogram ONLY)
max_freq (int): Only FFT bins corresponding to frequencies between
[0, max_freq] are returned (for spectrogram ONLY)
desc_file (str, optional): Path to a JSON-line file that contains
labels and paths to the audio files. If this is None, then
load metadata right away
"""
self.feat_dim = calc_feat_dim(window, max_freq)
self.mfcc_dim = mfcc_dim
self.feats_mean = np.zeros((self.feat_dim,))
self.feats_std = np.ones((self.feat_dim,))
self.rng = random.Random(RNG_SEED)
if desc_file is not None:
self.load_metadata_from_desc_file(desc_file)
self.step = step
self.window = window
self.max_freq = max_freq
self.cur_train_index = 0
self.cur_valid_index = 0
self.cur_test_index = 0
self.max_duration=max_duration
self.minibatch_size = minibatch_size
self.spectrogram = spectrogram
self.sort_by_duration = sort_by_duration
def get_batch(self, partition):
""" Obtain a batch of train, validation, or test data
"""
if partition == 'train':
audio_paths = self.train_audio_paths
cur_index = self.cur_train_index
texts = self.train_texts
elif partition == 'valid':
audio_paths = self.valid_audio_paths
cur_index = self.cur_valid_index
texts = self.valid_texts
elif partition == 'test':
audio_paths = self.test_audio_paths
cur_index = self.test_valid_index
texts = self.test_texts
else:
raise Exception("Invalid partition. "
"Must be train/validation")
features = [self.normalize(self.featurize(a)) for a in
audio_paths[cur_index:cur_index+self.minibatch_size]]
# calculate necessary sizes
max_length = max([features[i].shape[0]
for i in range(0, self.minibatch_size)])
max_string_length = max([len(texts[cur_index+i])
for i in range(0, self.minibatch_size)])
# initialize the arrays
X_data = np.zeros([self.minibatch_size, max_length,
self.feat_dim*self.spectrogram + self.mfcc_dim*(not self.spectrogram)])
labels = np.ones([self.minibatch_size, max_string_length]) * 28
input_length = np.zeros([self.minibatch_size, 1])
label_length = np.zeros([self.minibatch_size, 1])
for i in range(0, self.minibatch_size):
# calculate X_data & input_length
feat = features[i]
input_length[i] = feat.shape[0]
X_data[i, :feat.shape[0], :] = feat
# calculate labels & label_length
label = np.array(text_to_int_sequence(texts[cur_index+i]))
labels[i, :len(label)] = label
label_length[i] = len(label)
# return the arrays
outputs = {'ctc': np.zeros([self.minibatch_size])}
inputs = {'the_input': X_data,
'the_labels': labels,
'input_length': input_length,
'label_length': label_length
}
return (inputs, outputs)
def shuffle_data_by_partition(self, partition):
""" Shuffle the training or validation data
"""
if partition == 'train':
self.train_audio_paths, self.train_durations, self.train_texts = shuffle_data(
self.train_audio_paths, self.train_durations, self.train_texts)
elif partition == 'valid':
self.valid_audio_paths, self.valid_durations, self.valid_texts = shuffle_data(
self.valid_audio_paths, self.valid_durations, self.valid_texts)
else:
raise Exception("Invalid partition. "
"Must be train/validation")
def sort_data_by_duration(self, partition):
""" Sort the training or validation sets by (increasing) duration
"""
if partition == 'train':
self.train_audio_paths, self.train_durations, self.train_texts = sort_data(
self.train_audio_paths, self.train_durations, self.train_texts)
elif partition == 'valid':
self.valid_audio_paths, self.valid_durations, self.valid_texts = sort_data(
self.valid_audio_paths, self.valid_durations, self.valid_texts)
else:
raise Exception("Invalid partition. "
"Must be train/validation")
def next_train(self):
""" Obtain a batch of training data
"""
while True:
ret = self.get_batch('train')
self.cur_train_index += self.minibatch_size
if self.cur_train_index >= len(self.train_texts) - self.minibatch_size:
self.cur_train_index = 0
self.shuffle_data_by_partition('train')
yield ret
def next_valid(self):
""" Obtain a batch of validation data
"""
while True:
ret = self.get_batch('valid')
self.cur_valid_index += self.minibatch_size
if self.cur_valid_index >= len(self.valid_texts) - self.minibatch_size:
self.cur_valid_index = 0
self.shuffle_data_by_partition('valid')
yield ret
def next_test(self):
""" Obtain a batch of test data
"""
while True:
ret = self.get_batch('test')
self.cur_test_index += self.minibatch_size
if self.cur_test_index >= len(self.test_texts) - self.minibatch_size:
self.cur_test_index = 0
yield ret
def load_train_data(self, desc_file='train_corpus.json'):
self.load_metadata_from_desc_file(desc_file, 'train')
self.fit_train()
if self.sort_by_duration:
self.sort_data_by_duration('train')
def load_validation_data(self, desc_file='valid_corpus.json'):
self.load_metadata_from_desc_file(desc_file, 'validation')
if self.sort_by_duration:
self.sort_data_by_duration('valid')
def load_test_data(self, desc_file='test_corpus.json'):
self.load_metadata_from_desc_file(desc_file, 'test')
def load_metadata_from_desc_file(self, desc_file, partition):
""" Read metadata from a JSON-line file
(possibly takes long, depending on the filesize)
Params:
desc_file (str): Path to a JSON-line file that contains labels and
paths to the audio files
partition (str): One of 'train', 'validation' or 'test'
"""
audio_paths, durations, texts = [], [], []
with open(desc_file) as json_line_file:
for line_num, json_line in enumerate(json_line_file):
try:
spec = json.loads(json_line)
if float(spec['duration']) > self.max_duration:
continue
audio_paths.append(spec['key'])
durations.append(float(spec['duration']))
texts.append(spec['text'])
except Exception as e:
# Change to (KeyError, ValueError) or
# (KeyError,json.decoder.JSONDecodeError), depending on
# json module version
print('Error reading line #{}: {}'
.format(line_num, json_line))
if partition == 'train':
self.train_audio_paths = audio_paths
self.train_durations = durations
self.train_texts = texts
elif partition == 'validation':
self.valid_audio_paths = audio_paths
self.valid_durations = durations
self.valid_texts = texts
elif partition == 'test':
self.test_audio_paths = audio_paths
self.test_durations = durations
self.test_texts = texts
else:
raise Exception("Invalid partition to load metadata. "
"Must be train/validation/test")
def fit_train(self, k_samples=100):
""" Estimate the mean and std of the features from the training set
Params:
k_samples (int): Use this number of samples for estimation
"""
k_samples = min(k_samples, len(self.train_audio_paths))
samples = self.rng.sample(self.train_audio_paths, k_samples)
feats = [self.featurize(s) for s in samples]
feats = np.vstack(feats)
self.feats_mean = np.mean(feats, axis=0)
self.feats_std = np.std(feats, axis=0)
def featurize(self, audio_clip):
""" For a given audio clip, calculate the corresponding feature
Params:
audio_clip (str): Path to the audio clip
"""
if self.spectrogram:
return spectrogram_from_file(
audio_clip, step=self.step, window=self.window,
max_freq=self.max_freq)
else:
(rate, sig) = wav.read(audio_clip)
return mfcc(sig, rate, numcep=self.mfcc_dim)
def normalize(self, feature, eps=1e-14):
""" Center a feature using the mean and std
Params:
feature (numpy.ndarray): Feature to normalize
"""
return (feature - self.feats_mean) / (self.feats_std + eps)
def shuffle_data(audio_paths, durations, texts):
""" Shuffle the data (called after making a complete pass through
training or validation data during the training process)
Params:
audio_paths (list): Paths to audio clips
durations (list): Durations of utterances for each audio clip
texts (list): Sentences uttered in each audio clip
"""
p = np.random.permutation(len(audio_paths))
audio_paths = [audio_paths[i] for i in p]
durations = [durations[i] for i in p]
texts = [texts[i] for i in p]
return audio_paths, durations, texts
def sort_data(audio_paths, durations, texts):
""" Sort the data by duration
Params:
audio_paths (list): Paths to audio clips
durations (list): Durations of utterances for each audio clip
texts (list): Sentences uttered in each audio clip
"""
p = np.argsort(durations).tolist()
audio_paths = [audio_paths[i] for i in p]
durations = [durations[i] for i in p]
texts = [texts[i] for i in p]
return audio_paths, durations, texts
def vis_train_features(index=0):
""" Visualizing the data point in the training set at the supplied index
"""
# obtain spectrogram
audio_gen = AudioGenerator(spectrogram=True)
audio_gen.load_train_data()
vis_audio_path = audio_gen.train_audio_paths[index]
vis_spectrogram_feature = audio_gen.normalize(audio_gen.featurize(vis_audio_path))
# obtain mfcc
audio_gen = AudioGenerator(spectrogram=False)
audio_gen.load_train_data()
vis_mfcc_feature = audio_gen.normalize(audio_gen.featurize(vis_audio_path))
# obtain text label
vis_text = audio_gen.train_texts[index]
# obtain raw audio
vis_raw_audio, _ = librosa.load(vis_audio_path)
# print total number of training examples
print('There are %d total training examples.' % len(audio_gen.train_audio_paths))
# return labels for plotting
return vis_text, vis_raw_audio, vis_mfcc_feature, vis_spectrogram_feature, vis_audio_path
def plot_raw_audio(vis_raw_audio):
# plot the raw audio signal
fig = plt.figure(figsize=(12,3))
ax = fig.add_subplot(111)
steps = len(vis_raw_audio)
ax.plot(np.linspace(1, steps, steps), vis_raw_audio)
plt.title('Audio Signal')
plt.xlabel('Time')
plt.ylabel('Amplitude')
plt.show()
def plot_mfcc_feature(vis_mfcc_feature):
# plot the MFCC feature
fig = plt.figure(figsize=(12,5))
ax = fig.add_subplot(111)
im = ax.imshow(vis_mfcc_feature, cmap=plt.cm.jet, aspect='auto')
plt.title('Normalized MFCC')
plt.ylabel('Time')
plt.xlabel('MFCC Coefficient')
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="5%", pad=0.05)
plt.colorbar(im, cax=cax)
ax.set_xticks(np.arange(0, 13, 2), minor=False);
plt.show()
def plot_spectrogram_feature(vis_spectrogram_feature):
# plot the normalized spectrogram
fig = plt.figure(figsize=(12,5))
ax = fig.add_subplot(111)
im = ax.imshow(vis_spectrogram_feature, cmap=plt.cm.jet, aspect='auto')
plt.title('Normalized Spectrogram')
plt.ylabel('Time')
plt.xlabel('Frequency')
divider = make_axes_locatable(ax)
cax = divider.append_axes("right", size="5%", pad=0.05)
plt.colorbar(im, cax=cax)
plt.show()