-
Notifications
You must be signed in to change notification settings - Fork 2
/
extract_test_feature.py
328 lines (273 loc) · 13 KB
/
extract_test_feature.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
import os
import csv
import sys
import math
import cv2
import scipy
import pickle
import librosa
import matplotlib
import numpy as np
import librosa.display
import IPython.display as ipd
matplotlib.use('TkAgg')
import matplotlib.pyplot as plt
# print(os.getcwd())
# sys.exit(1)
# Import other project files
from read_test_data import get_non_prog_files,get_prog_files
sys.setrecursionlimit(10000)
prog_files = get_prog_files()
non_prog_files = get_non_prog_files()
print("Number of prog songs",len(prog_files))
print("Number of non prog songs",len(non_prog_files))
all_files = prog_files + non_prog_files
fixed_sr = 44100
min_duration = 0
# -------------------------- Find minimum duration -------------------------------------------------
'''
for i in range(len(all_files)) :
filename = all_files[i]
y, sr = librosa.load(filename,sr = fixed_sr)
curr_duration = librosa.get_duration(y=y, sr=sr)
if i == 0 :
min_duration = curr_duration
else :
min_duration = min(min_duration,curr_duration)
'''
min_duration = 30#60.041
min_duration = int(min_duration)
print("min duration ",min_duration)
# -------------------------- Feature extraction ----------------------------------------------------
file = open('test_features.csv', 'w', newline='')
header = 'filename genre chroma_stft rmse spectral_centroid spectral_bandwidth rolloff zero_crossing_rate'
for i in range(1, 21):
header += f' mfcc{i}'
header = header.split()
# Create file to write error logs
error_logs = open("error_logs.txt","w")
error_logs.close()
with file:
writer = csv.writer(file)
writer.writerow(header)
genre = 'prog'
count = 0
time_series_length = 30
# Read prog files
for i in range(len(prog_files)) :
print("prog ",i)
filename = prog_files[i]
name = (filename.split("/") )[-1]
name = name.replace(" ","_")
try:
y, sr = librosa.load(filename,sr = fixed_sr)
time = librosa.get_duration(y=y,sr=sr)
chunks = []
if time > min_duration :
org_y = y
iter = math.floor(time/min_duration)
print(iter)
current_size = time*fixed_sr
chunk_size = min_duration*fixed_sr
start = 0#math.floor(iter/3) * chunk_size
end = chunk_size
chunk_index = 1
# iter = math.floor(iter/3)
while iter !=0 :
count += 1
chunk = y[start:end]
chroma_stft = librosa.feature.chroma_stft(y=chunk, sr=sr)
spec_cent = librosa.feature.spectral_centroid(y=chunk, sr=sr)
spec_bw = librosa.feature.spectral_bandwidth(y=chunk, sr=sr)
rmse = librosa.feature.rmse(y=chunk)
rolloff = librosa.feature.spectral_rolloff(y=chunk, sr=sr)
zcr = 10**10*np.mean(librosa.zero_crossings(org_y)/len(org_y) )
mfcc = librosa.feature.mfcc(y=chunk, sr=sr)
if np.mean(chroma_stft) == 0 :
break
to_append = f'{"prog"+name+"chunk"+str(chunk_index)} {genre} {np.mean(chroma_stft)} {np.mean(rmse)} {np.mean(spec_cent)} {np.mean(spec_bw)} {np.mean(rolloff)} {zcr}'
# Append all mfcc features i.e., 20 rows
for e in mfcc:
to_append += f' {np.mean(e)}'
file = open('test_features.csv', 'a', newline='')
with file:
writer = csv.writer(file)
writer.writerow(to_append.split())
# spect = librosa.feature.melspectrogram(y=chunk, sr=sr,n_fft=2048, hop_length=512)
# spect = librosa.power_to_db(spect, ref=np.max)
# plt.figure(figsize=(14, 5))
# plt.axis('off')
# librosa.display.specshow(spect, fmax=8000)
# plt.savefig("outputprog.png",bbox_inches='tight',transparent=True,pad_inches=0)
# im = cv2.imread("outputprog.png")
# plt.clf()
# plt.cla()
# plt.close()
# np.savez_compressed("/Users/richadutt/Documents/ranga/mfcc_validation_set/prog-"+name+"chunk"+str(chunk_index),im,im)
chunk_index += 1
start = end
end = end + chunk_size
iter -= 1
# if chunk_index >=20 :
# break
# print("chunk size ",chunk_size)
# print("current song size ",current_size)
else :
count += 1
y, sr = librosa.load(filename,sr = fixed_sr,duration=min_duration)
chroma_stft = librosa.feature.chroma_stft(y=y, sr=sr)
spec_cent = librosa.feature.spectral_centroid(y=y, sr=sr)
spec_bw = librosa.feature.spectral_bandwidth(y=y, sr=sr)
rmse = librosa.feature.rmse(y=y)
rolloff = librosa.feature.spectral_rolloff(y=y, sr=sr)
zcr = 10**10*np.mean(librosa.zero_crossings(org_y)/len(org_y) )
mfcc = librosa.feature.mfcc(y=y, sr=sr)
to_append = f'{"prog"+name+"chunk1"} {genre} {np.mean(chroma_stft)} {np.mean(rmse)} {np.mean(spec_cent)} {np.mean(spec_bw)} {np.mean(rolloff)} {zcr}'
# Append all mfcc features i.e., 20 rows
for e in mfcc:
to_append += f' {np.mean(e)}'
file = open('test_features.csv', 'a', newline='')
with file:
writer = csv.writer(file)
writer.writerow(to_append.split())
# spect = librosa.feature.melspectrogram(y=y, sr=sr,n_fft=2048, hop_length=512)
# spect = librosa.power_to_db(spect, ref=np.max)
# plt.figure(figsize=(14, 5))
# plt.axis('off')
# X = librosa.stft(y)
# Xdb = librosa.amplitude_to_db(abs(X))
# librosa.display.specshow(Xdb, sr=sr, x_axis='time', y_axis='hz')
# plt.savefig("outputprog.png",bbox_inches='tight',transparent=True,pad_inches=0)
# im = cv2.imread("outputprog.png")
# plt.clf()
# plt.cla()
# plt.close()
# np.savez_compressed("/Users/richadutt/Documents/ranga/mfcc_validation_set/prog-"+name+"chunk1",im,im)
# print(i," ----> ",y.shape)
except Exception as e :
print("error handled")
error_logs = open("error_logs.txt","a")
error_logs.write(filename)
error_logs.write("\n")
error_logs.write(str(e))
error_logs.write("\n")
error_logs.close()
continue
genre = 'non_prog'
# Read Non prog files
for i in range(len(non_prog_files)) :
print("nonprog ",i)
filename = non_prog_files[i]
name = (filename.split("/") )[-1]
name = name.replace(" ","_")
# print("target ",duration*fixed_sr)
try:
y, sr = librosa.load(filename,sr = fixed_sr)
time = librosa.get_duration(y=y,sr=sr)
chunks = []
if time > min_duration :
org_y = y
iter = math.floor(time/min_duration)
print(iter)
current_size = time*fixed_sr
chunk_size = min_duration*fixed_sr
start = 0#math.floor(iter/3) * chunk_size
end = chunk_size
chunk_index = 1
# iter = math.floor(iter/3)
while iter !=0 :
count += 1
chunk = y[start:end]
# spect = librosa.feature.melspectrogram(y=chunk, sr=sr,n_fft=2048, hop_length=512)
# spect = librosa.power_to_db(spect, ref=np.max)
# plt.figure(figsize=(14, 5))
# plt.axis('off')
# X = librosa.stft(chunk)
# Xdb = librosa.amplitude_to_db(abs(X))
# librosa.display.specshow(Xdb, sr=sr, x_axis='time', y_axis='hz')
# plt.savefig("outputnonprog.png",bbox_inches='tight',transparent=True,pad_inches=0)
# im = cv2.imread("outputnonprog.png")
# plt.clf()
# plt.cla()
# plt.close()
chroma_stft = librosa.feature.chroma_stft(y=chunk, sr=sr)
spec_cent = librosa.feature.spectral_centroid(y=chunk, sr=sr)
spec_bw = librosa.feature.spectral_bandwidth(y=chunk, sr=sr)
rmse = librosa.feature.rmse(y=chunk)
rolloff = librosa.feature.spectral_rolloff(y=chunk, sr=sr)
zcr = 10**10*np.mean(librosa.zero_crossings(org_y)/len(org_y) )
mfcc = librosa.feature.mfcc(y=chunk, sr=sr)
if np.mean(chroma_stft) == 0 :
break
to_append = f'{"nonprog"+name+"chunk"+str(chunk_index)} {genre} {np.mean(chroma_stft)} {np.mean(rmse)} {np.mean(spec_cent)} {np.mean(spec_bw)} {np.mean(rolloff)} {zcr}'
# Append all mfcc features i.e., 20 rows
for e in mfcc:
to_append += f' {np.mean(e)}'
file = open('test_features.csv', 'a', newline='')
with file:
writer = csv.writer(file)
writer.writerow(to_append.split())
# np.savez_compressed("/Users/richadutt/Documents/ranga/mfcc_validation_set/nonprog-"+name+"chunk"+str(chunk_index),im,im)
chunk_index += 1
# print("start ",start)
# print("end ",end)
start = end
end = end + chunk_size
iter -= 1
# if chunk_index >=20 :
# break
# working 4 mailed 4 now 10
# print("chunk size ",chunk_size)
# print("current song size ",current_size)
else :
count += 1
y, sr = librosa.load(filename,sr = fixed_sr,duration=min_duration)
# spect = librosa.feature.melspectrogram(y=y, sr=sr,n_fft=2048, hop_length=512)
# spect = librosa.power_to_db(spect, ref=np.max)
# plt.figure(figsize=(14, 5))
# plt.axis('off')
# X = librosa.stft(y)
# Xdb = librosa.amplitude_to_db(abs(X))
# librosa.display.specshow(Xdb, sr=sr, x_axis='time', y_axis='hz')
# plt.savefig("outputnonprog.png",bbox_inches='tight',transparent=True,pad_inches=0)
# im = cv2.imread("outputnonprog.png")
# plt.clf()
# plt.cla()
# plt.close()
chroma_stft = librosa.feature.chroma_stft(y=y, sr=sr)
spec_cent = librosa.feature.spectral_centroid(y=y, sr=sr)
spec_bw = librosa.feature.spectral_bandwidth(y=y, sr=sr)
rmse = librosa.feature.rmse(y=y)
rolloff = librosa.feature.spectral_rolloff(y=y, sr=sr)
zcr = 10**10*np.mean(librosa.zero_crossings(org_y)/len(org_y) )
mfcc = librosa.feature.mfcc(y=y, sr=sr)
to_append = f'{"nonprog"+name+"chunk1"} {genre} {np.mean(chroma_stft)} {np.mean(rmse)} {np.mean(spec_cent)} {np.mean(spec_bw)} {np.mean(rolloff)} {zcr}'
# Append all mfcc features i.e., 20 rows
for e in mfcc:
to_append += f' {np.mean(e)}'
file = open('test_features.csv', 'a', newline='')
with file:
writer = csv.writer(file)
writer.writerow(to_append.split())
# np.savez_compressed("/Users/richadutt/Documents/ranga/mfcc_validation_set/nonprog-"+name+"chunk1",im,im)
except Exception as e:
print("error handled")
error_logs = open("error_logs.txt","a")
error_logs.write(filename)
error_logs.write("\n")
error_logs.write(str(e))
error_logs.write("\n")
error_logs.close()
continue
print("tot count ",count)
# Display Spectrogram
# STFT converts signal such that we can know the amplitude of given frequency at a given time.
# Using STFT we can determine the amplitude of various frequencies playing at a given time of an audio signal
'''
X = librosa.stft(x)
Xdb = librosa.amplitude_to_db(abs(X))
plt.figure(figsize=(14, 5))
librosa.display.specshow(Xdb, sr=sr, x_axis='time', y_axis='hz')
plt.colorbar()
plt.show()
'''