def parse_audio_w2l2(self, audio_path):
if self.augment:
y = (load_randomly_augmented_audio(audio_path, self.sample_rate,w2l2=True)).astype(np.float32)
else:
y = load_audio(audio_path)
if self.noiseInjector:
add_noise = np.random.binomial(1, self.noise_prob)
if add_noise:
y = self.noiseInjector.inject_noise(y)
n_fft = 512
mfcc = pytorch_mfcc.MFCC(samplerate=self.sample_rate,winlen=self.window_size,winstep=self.window_stride,numcep=13,nfilt=26,nfft=n_fft,lowfreq=0,highfreq=None,preemph=0,ceplifter=22,appendEnergy=False).cuda()
y = Variable(y, requires_grad=True)
mfccs = mfcc(y)
mean = mfccs.mean()
std = mfccs.std()
mfccs = torch.add(mfccs, -mean)
mfccs = mfccs / std
return mfccs.data
def attack(self, iterations, target_path, lr, bandwidth):
flag = 0
model = WaveToLetter.load_model(args.model_path)
model = model.to(device)
model.eval() #eval. This is different from stage1
signal = self.get_signal(self.audio_path)
orig = self.get_signal(self.orig_path)
index_max, index_min = self.attention(signal)
start_attack_time = time.time()
for i in range(iterations):
print('Iteration:', str(i))
# print(signal[index_max:index_max+20])
# print(signal[index_max])
# if args.printSilence:
# print()
# print(sigindex_maxnal.shape)
# print("20：", signal[index_max:index_max+20])
mfcc = pytorch_mfcc.MFCC(samplerate=self.sample_rate,
winlen=self.window_size,
winstep=self.window_stride,
numcep=13,
nfilt=26,
nfft=512,
lowfreq=0,
highfreq=None,
preemph=0,
ceplifter=22,
appendEnergy=False).cuda()
mfccs = mfcc(signal)
mfccs = self.normalize(mfccs)
if args.printSilence:
print("mfccs", mfccs)
inputsMags = self.mfccs_to_inputs(mfccs)
out = model(inputsMags)
path = self.orig_path.split(
'wav')[0] + 'txt' + self.orig_path.split('wav')[1]
fp = open(path)
transcriptReal = fp.readlines()[0]
print("Ref:", transcriptReal.lower())
seq_length = out.size(1)
sizes = Variable(torch.Tensor([1.0]).mul_(int(seq_length)).int(),
requires_grad=False)
if args.printSilence:
print("out", out)
print("softmax", F.softmax(out, dim=-1).data)
decoded_output, _, = decoder.decode(
F.softmax(out, dim=-1).data, sizes)
transcript = decoded_output[0][0]
print("Hyp:", transcript.lower())
out = out.transpose(0, 1)
if args.target:
transcriptTarget = args.target
else:
fp = open(target_path)
transcriptTarget = fp.readlines()[0]
print("Tar:", transcriptTarget.lower())
if transcript.lower() == transcriptTarget.lower() and i > 0:
if args.target:
target_path = args.target
save_path = self.save(signal, target_path, self.orig_path, lr,
iterations, i, bandwidth)
generate_time = time.time() - start_attack_time
print('Time taken (s): {generate_time:.4f}\t'.format(
generate_time=generate_time))
self.save_figure(signal, save_path)
break
target = list(
filter(None, [
self.labels_map.get(x)
for x in list(transcriptTarget.upper())
]))
targets = torch.IntTensor(target)
target_sizes = torch.IntTensor([len(target)])
ctcloss = self.criterion(out, targets, sizes, target_sizes)
# print("ctcloss:", ctcloss)
# print("delta_2:", 100*torch.sum((signal - orig)**2))
# loss = ctcloss + 100*torch.sum((signal - orig)**2)
loss = ctcloss
print("loss:", loss)
loss.backward()
grad = np.array(signal.grad)
is_nan = np.isnan(grad)
is_nan_new = is_nan[is_nan == True]
for j in range(len(grad)):
if is_nan[j]:
grad[j] = 10
wer = decoder.wer(transcript.lower(),
transcriptTarget.lower()) / float(
len(transcriptTarget.lower().split()))
# the iterative proportional clipping method
# print('grad:{}'.format(grad[index_max:index_max+20]))
perturbation = lr * torch.from_numpy(grad)
# print('perturbation', perturbation[index_max])
signal_next_relative = torch.clamp(
(signal.data - perturbation) / orig,
min=1 - bandwidth,
max=1 + bandwidth)
# print("signal_next_relative1:", signal_next_relative[index_max])
signal.data = signal_next_relative.mul(orig)
# print(signal_next_relative[index_max]*orig[index_max])
# print("signal.data:", signal.data[index_max])
# if (i + 1) % 15000 == 0 and flag < 1:
# # anneal lr
# # lr *= 0.5
# lr = lr / args.learning_anneal
# flag += 1
# print("wer", wer)
# print("lr", lr)
print("\n")
signal.grad.data.zero_()
print("Come to the end")