def load_snore(self):
snores = []
wavpaths = [os.path.join(self.snorepath,i) for i in os.listdir(self.snorepath) if '.wav' in i]
for wavpath in wavpaths:
wavsignal, fs = read_wav_data(wavpath)
wavsignal = wavsignal[0]
snores.append( (wavsignal, wavpath) )
return snores
def load_one_noise(self, noise_type):
noise_segs = []
paths = self.__load_one_noise_paths__(noise_type)
# self.number_check(paths)
for apath in paths:
seg_path = os.path.join(self.setpath, apath[0], apath[1], apath[2])
wavsignal, fs = read_wav_data(seg_path)
wavsignal = wavsignal[0]
output = (wavsignal, seg_path)
noise_segs.append(output)
return noise_segs
def GetData(self, n_start, n_amount=32, dataType = 'same'):
assert(n_amount%CLASS_NUM==0)
if dataType == self.pathSameLabel:
path = self.listSame[n_start][0]
data_label = np.array([self.listSame[n_start][1]])
elif dataType == self.pathDistinctLabel:
path = self.listDistinct[n_start][0]
data_label = np.array([self.listDistinct[n_start][1]])
wavsignal, fs = read_wav_data(path)
if FEATURE_TYPE in ['spec', 'Spec', 'SPEC', 'Spectrogram', 'SPECTROGRAM']:
data_input = GetFrequencyFeatures(wavsignal, fs, self.feat_dimension, self.frame_length, shift=160)
if PRIOR_ART == True:
data_input = hog(data_input, orientations=8, pixels_per_cell=(16, 16), cells_per_block=(1, 1),visualise=False)
return data_input, data_label
elif FEATURE_TYPE in ['mfcc', 'MFCC', 'Mfcc']:
data_input = SimpleMfccFeatures(wavsignal, fs)
else:
print('Unknown feature type.')
assert (0)
data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
return data_input, data_label
import os
from feature_transformation import read_wav_data, GetFrequencyFeatures
import matplotlib.pyplot as plt
if __name__ == '__main__':
apath = '../../Data/BSfilteringData/data/762758/bowels/81730697302190_2018_12_20_12_02_42.wav'
wavsignal, fs = read_wav_data(apath)
wavsignal = wavsignal[0]
plt.plot(wavsignal)
plt.show()
def GetData(self, n_start, n_amount=32, featureType = 'spectrogram', mode='balanced'):
'''
读取数据，返回神经网络输入值和输出值矩阵(可直接用于神经网络训练的那种)
参数：
n_start：从编号为n_start数据开始选取数据
n_amount：选取的数据数量，默认为1，即一次一个wav文件
返回：
四个音频四个label，
'''
# 随机从四个文件夹中拿一条数据，判断是否大于1s，否就重拿
assert(n_amount%CLASS_NUM==0)
category = (self.list_bowel1, self.list_non0)
link = ('bowels', 'non')
label = (1, 0)
# extract = {'spectrogram':GetFrequencyFeatures,'mfcc':mfccFeatures}
path = ''
if FEATURE_TYPE in ['spec','Spec','SPEC','Spectrogram','SPECTROGRAM']:
if mode == 'balanced':
data = []
labels = []
for genre in range(CLASS_NUM):
for file in range(n_amount//CLASS_NUM):
filename = category[genre][(n_start + file)%self.DataNum[genre]]
# filename = category[genre][(n_start + file) % min(self.DataNum)]
path = self.common_path + link[genre] + self.slash + filename
wavsignal, fs = read_wav_data(path)
# data_input = SimpleMfccFeatures(wavsignal, fs)
data_input = GetFrequencyFeatures(wavsignal, fs, self.feat_dimension, self.frame_length,shift=160)
data_input = self.shifting(data_input)
if PRIOR_ART == True:
data_input = hog(data_input, orientations=8, pixels_per_cell=(16, 16), cells_per_block=(1, 1), visualise=False)
# data_input = MelSpectrogram(wavsignal, fs,frame_length = self.frame_length, shift=160,filternum = 26)
else:
data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
data.append(data_input)
data_label = np.array([label[genre]])
labels.append(data_label)
return data, labels
if mode == 'non-repetitive':
path = self.list_path[n_start][0]
data_label = np.array([self.list_path[n_start][1]])
wavsignal, fs = read_wav_data(path)
data_input = GetFrequencyFeatures(wavsignal, fs, self.feat_dimension, self.frame_length,shift=160)
# data_input = MelSpectrogram(wavsignal, fs, frame_length=self.frame_length, shift=160, filternum=26)
# data_input = SimpleMfccFeatures(wavsignal, fs)
if PRIOR_ART == True:
data_input = hog(data_input, orientations=8, pixels_per_cell=(16, 16), cells_per_block=(1, 1),visualise=False)
else:
data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
return data_input, data_label
elif FEATURE_TYPE in ['mfcc','MFCC','Mfcc']:
if mode == 'balanced':
data = []
labels = []
for genre in range(CLASS_NUM):
for file in range(n_amount//CLASS_NUM):
filename = category[genre][(n_start + file)%self.DataNum[genre]]
# filename = category[genre][(n_start + file) % min(self.DataNum)]
path = self.common_path + link[genre] + self.slash + filename
wavsignal, fs = read_wav_data(path)
data_input = SimpleMfccFeatures(wavsignal, fs)
data_label = np.array([label[genre]])
# if data_label[0] == 0:
# if PRIOR_ART == False:
data_input = self.shifting(data_input)
data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
data.append(data_input)
labels.append(data_label)
return data, labels
if mode == 'non-repetitive':
path = self.list_path[n_start][0]
data_label = np.array([self.list_path[n_start][1]])
wavsignal, fs = read_wav_data(path)
data_input = SimpleMfccFeatures(wavsignal, fs)
data_input = data_input.reshape(data_input.shape[0], data_input.shape[1], 1)
return data_input, data_label
else:
print('Unknown feature type.')
assert(0)
