def plot_audio(full_file_path, plot_type, radioIntVar):
spf = wave.open(full_file_path, 'r')
#Extract Raw Audio from Wav File
opened_signal = spf.readframes(-1)
opened_signal = np.fromstring(opened_signal, 'Int16')
#opened_signal = np.array_split(opened_signal, Plot.sub_arrays)
global fs
fs = spf.getframerate()
#If Stereo
if spf.getnchannels() == 2:
print 'Just mono files'
sys.exit(0)
if (radioIntVar.get() == 1): # is 2D plot
if (plot_type == "raw"):
Plot.plot_raw_audio2D(opened_signal)
elif (plot_type == "fft"):
Plot.plot_fft_audio2D(opened_signal)
elif (plot_type == "spectrogram"):
#Plot.plot_spectrogram2D(opened_signal)
plotstft(full_file_path, generatefig=False)
elif (radioIntVar.get() == 2): #multi dimensional plot
opened_signal = np.array_split(opened_signal, Plot.sub_arrays)
if (plot_type == "raw"):
Plot.plot_raw_audio(opened_signal)
elif (plot_type == "fft"):
Plot.plot_fft_audio(opened_signal)
def plot_audio(full_file_path, plot_type, radioIntVar):
spf = wave.open(full_file_path,'r')
#Extract Raw Audio from Wav File
opened_signal = spf.readframes(-1)
opened_signal = np.fromstring(opened_signal, 'Int16')
#opened_signal = np.array_split(opened_signal, Plot.sub_arrays)
global fs
fs = spf.getframerate()
#If Stereo
if spf.getnchannels() == 2:
print 'Just mono files'
sys.exit(0)
if(radioIntVar.get() == 1): # is 2D plot
if(plot_type == "raw"):
Plot.plot_raw_audio2D(opened_signal)
elif(plot_type == "fft"):
Plot.plot_fft_audio2D(opened_signal)
elif(plot_type == "spectrogram"):
#Plot.plot_spectrogram2D(opened_signal)
plotstft(full_file_path, generatefig=False)
elif(radioIntVar.get() == 2): #multi dimensional plot
opened_signal = np.array_split(opened_signal, Plot.sub_arrays)
if(plot_type == "raw"):
Plot.plot_raw_audio(opened_signal)
elif(plot_type == "fft"):
Plot.plot_fft_audio(opened_signal)
def predict_results():
# generating graph
spectogram.plotstft('test.wav', generatefig=True) # display user recorded sound and save plot
test_img = cv2.imread('test.png', 0) # load grayscale pic
cv2.imshow('test.png', test_img) # display new pic
X_test = [] # preparing input
X_test.append(test_img)
X_test = np.array(X_test)
X_test = X_test.reshape(X_test.shape[0], 1, 33, 70) # 33 num of rows, 70 num of columns
# testing input
results_test = NeuralNetwork.ann.predict_classes(X_test, batch_size=1)
result_probability = NeuralNetwork.ann.predict_proba(X_test, batch_size=1)
result_probability *= 100 # convert to percent
print "[Winner]: " + NeuralNetwork.alphabet[results_test]
print "[Probabilities:] " + "ASC:" + str(result_probability[0, 0])+"%" + ", DESC:" + str(result_probability[0, 1])+"%" + \
", FLAT:" + str(result_probability[0, 2])+"%" + ", SOY: " + str(result_probability[0, 3])+"%"
def predict_results():
# generating graph
spectogram.plotstft(
'test.wav',
generatefig=True) # display user recorded sound and save plot
test_img = cv2.imread('test.png', 0) # load grayscale pic
cv2.imshow('test.png', test_img) # display new pic
X_test = [] # preparing input
X_test.append(test_img)
X_test = np.array(X_test)
X_test = X_test.reshape(X_test.shape[0], 1, 33,
70) # 33 num of rows, 70 num of columns
# testing input
results_test = NeuralNetwork.ann.predict_classes(X_test, batch_size=1)
result_probability = NeuralNetwork.ann.predict_proba(X_test,
batch_size=1)
result_probability *= 100 # convert to percent
print "[Winner]: " + NeuralNetwork.alphabet[results_test]
print "[Probabilities:] " + "ASC:" + str(result_probability[0, 0])+"%" + ", DESC:" + str(result_probability[0, 1])+"%" + \
", FLAT:" + str(result_probability[0, 2])+"%" + ", SOY: " + str(result_probability[0, 3])+"%"
