def getCepsVect(y):
ceps, mspec, spec = mfcc(y)
vect_of_mccf = np.zeros(COEFS)
for j in range(COEFS):
vect_of_mccf[j] = RecordModule.arithmeticMean(ceps.T[j])
vect_of_mccf[0] = 0
return vect_of_mccf
def getCepsMatrixFromData(t,y):
'''
funkcja wyznacz wspolczynniki MFCC z danych otrzymanych jako parametry
t: czas podany jako tablica kolejnych wartosci
y: sygnal podany jako tablica wartosci
'''
##########
y = sigproc.preemp(y)
fr, wordspower, wordszeros, wordsdetect, ITL ,ITU, word_fr, word_y = RecordModule.detectSingleWord(t,y)
##################
MelFeat = mealfeat.MelFeatures()
ceps_matrix = MelFeat.calcMelMatrixFeatures(word_y)
return ceps_matrix
def goRecognition():
'''
funkcja nagrywa komendem, a nastepnie klasyfikuje ja do odpowiedniej klasy
'''
print("please speak a word into the microphone")
t, y = RecordModule.getSpeechFromMic()
print("done")
print("dl probki :",len(y))
# t,y = PlotModule.readWav("learn_set//wylacz//9.wav", 44100.0)
predict = getCepsMatrixFromData(t, y)
predClass = getClasificationDecision(predict)
if(MODE == 1):
SendCommandToSerialPort(predClass.name)
print("done")
ylabel('|Y(freq)|')
def readWav(filename, Fs):
'''
funkcja czyta plik dzwiekowy
'''
rate, data = read(filename)
y = numpy.atleast_2d(data)[0]
lungime = len(y)
timp = len(y) / Fs
t = linspace(0, timp, len(y))
return t, y
if __name__ == '__main__':
Fs = 44100.0
# sampling rate
filename = "learn_set//wlacz//6.wav"
t, y = readWav(filename, Fs)
y = sigproc.preemp(y, 0.97)
fr, wordspower, wordszeros, wordsdetect, ITL, ITU, word_fr, word_y = RecordModule.detectSingleWord(
t, y)
pylab.subplot(211)
pylab.title(filename)
pylab.plot(t, y)
pylab.subplot(212)
print(word_y.shape)
plotSpectrum(word_y, Fs)
pylab.show()
return vect_of_mccf
if __name__ == '__main__':
# ++++++++++++++++++++++++++++++++++++++++++++
for i in range(11):
print("please speak a word into the microphone")
filename = "learn_set//wlacz//"+str(i+1)+".wav"
# RecordModule.record_to_file(filename)
# print("done - result written to ", filename)
# filename = 'learn_set//wlacz//3.wav'
# ++++++++++++++++++++++++++++++++++++++++++++
t, extract = PlotModule.readWav(filename, FS)
extract = RecordModule.preemp(extract)
fr, wordspower, wordszeros, wordsdetect, ITL ,ITU, word_fr, word_y = RecordModule.detectSingleWord(t,extract)
ceps, mspec, spec = mfcc(word_y)
show_MFCC_spectrum(ceps)
# show_MFCC(ceps)
vect_of_mccf = np.zeros(len(ceps.T))
for i in range(len(ceps.T)):
vect_of_mccf[i] = max(ceps.T[i]) # sum(data.T[i]) #
# Compute the spectrum magnitude
spec = np.abs(fft(framed, nfft, axis=-1))
# Filter the spectrum through the triangle filterbank
mspec = np.log10(np.dot(spec, fbank.T))
# Use the DCT to 'compress' the coefficients (spectrum -> cepstrum domain)
ceps = dct(mspec, type=2, norm='ortho', axis=-1)[:, :nceps]
return ceps, mspec, spec
def preemp(input, p):
"""Pre-emphasis filter."""
return lfilter([1., -p], 1, input)
if __name__ == '__main__':
for i in range(10):
filename = "learn_set//wlacz//"+str(i+1)+".wav"
RATE = 44100.0
t,y = PlotModule.readWav(filename, RATE)
ceps, mspec, spec = mfcc(y)
print(ceps.shape)
vect_of_mccf = np.zeros(len(ceps.T))
for i in range(len(ceps.T)):
vect_of_mccf[i] = RecordModule.arithmeticMean(ceps.T[i]) # sum(data.T[i]) #
pylab.title("ceps : ")
pylab.plot(range(len(vect_of_mccf)), vect_of_mccf, 'g')
pylab.show()
mspec = np.log10(np.dot(spec, fbank.T))
# Use the DCT to 'compress' the coefficients (spectrum -> cepstrum domain)
ceps = dct(mspec, type=2, norm='ortho', axis=-1)[:, :nceps]
return ceps, mspec, spec
def preemp(input, p):
"""Pre-emphasis filter."""
return lfilter([1., -p], 1, input)
if __name__ == '__main__':
for i in range(10):
filename = "learn_set//wlacz//" + str(i + 1) + ".wav"
RATE = 44100.0
t, y = PlotModule.readWav(filename, RATE)
ceps, mspec, spec = mfcc(y)
print(ceps.shape)
vect_of_mccf = np.zeros(len(ceps.T))
for i in range(len(ceps.T)):
vect_of_mccf[i] = RecordModule.arithmeticMean(
ceps.T[i]) # sum(data.T[i]) #
pylab.title("ceps : ")
pylab.plot(range(len(vect_of_mccf)), vect_of_mccf, 'g')
pylab.show()
plot(frq,abs(Y),'r') # plotting the spectrum
xlabel('Freq (Hz)')
ylabel('|Y(freq)|')
def readWav(filename, Fs):
'''
funkcja czyta plik dzwiekowy
'''
rate,data=read(filename)
y = numpy.atleast_2d(data)[0]
lungime=len(y)
timp=len(y)/Fs
t=linspace(0,timp,len(y))
return t, y
if __name__ == '__main__':
Fs = 44100.0; # sampling rate
filename = "learn_set//wlacz//6.wav"
t,y = readWav(filename, Fs)
y = sigproc.preemp(y,0.97)
fr, wordspower, wordszeros, wordsdetect, ITL ,ITU, word_fr, word_y = RecordModule.detectSingleWord(t,y)
pylab.subplot(211)
pylab.title(filename)
pylab.plot(t, y)
pylab.subplot(212)
print(word_y.shape)
plotSpectrum(word_y,Fs)
pylab.show()
