def test_mu_law_companding(self):
sig = self.sig.clone()
quantization_channels = 256
sig = self.sig.numpy()
sig = sig / np.abs(sig).max()
self.assertTrue(sig.min() >= -1. and sig.max() <= 1.)
sig_mu = transforms.MuLawEncoding(quantization_channels)(sig)
self.assertTrue(sig_mu.min() >= 0.
and sig.max() <= quantization_channels)
sig_exp = transforms.MuLawExpanding(quantization_channels)(sig_mu)
self.assertTrue(sig_exp.min() >= -1. and sig_exp.max() <= 1.)
#diff = sig - sig_exp
#mse = np.linalg.norm(diff) / diff.shape[0]
#self.assertTrue(mse, np.isclose(mse, 0., atol=1e-4)) # not always true
sig = self.sig.clone()
sig = sig / torch.abs(sig).max()
self.assertTrue(sig.min() >= -1. and sig.max() <= 1.)
sig_mu = transforms.MuLawEncoding(quantization_channels)(sig)
self.assertTrue(sig_mu.min() >= 0.
and sig.max() <= quantization_channels)
sig_exp = transforms.MuLawExpanding(quantization_channels)(sig_mu)
self.assertTrue(sig_exp.min() >= -1. and sig_exp.max() <= 1.)
def test_mu_law_companding(self):
sig = self.sig.clone()
quantization_channels = 256
sig = self.sig.numpy()
sig = sig / np.abs(sig).max()
self.assertTrue(sig.min() >= -1. and sig.max() <= 1.)
sig_mu = transforms.MuLawEncoding(quantization_channels)(sig)
self.assertTrue(sig_mu.min() >= 0. and sig.max() <= quantization_channels)
sig_exp = transforms.MuLawExpanding(quantization_channels)(sig_mu)
self.assertTrue(sig_exp.min() >= -1. and sig_exp.max() <= 1.)
sig = self.sig.clone()
sig = sig / torch.abs(sig).max()
self.assertTrue(sig.min() >= -1. and sig.max() <= 1.)
sig_mu = transforms.MuLawEncoding(quantization_channels)(sig)
self.assertTrue(sig_mu.min() >= 0. and sig.max() <= quantization_channels)
sig_exp = transforms.MuLawExpanding(quantization_channels)(sig_mu)
self.assertTrue(sig_exp.min() >= -1. and sig_exp.max() <= 1.)
repr_test = transforms.MuLawEncoding(quantization_channels)
repr_test.__repr__()
repr_test = transforms.MuLawExpanding(quantization_channels)
repr_test.__repr__()
targets = targets[:x[-1].astype(int)]
inputs = inputs[:len(inputs) // seq_M * seq_M]
targets = targets[:len(targets) // seq_M * seq_M]
h = f(np.arange(1, len(inputs) + 1))
train_wav.append(inputs)
train_features.append(h)
train_targets.append(targets)
train_wav = np.concatenate(train_wav)
train_features = np.vstack(train_features)
train_targets = np.concatenate(train_targets)
enc = transforms.MuLawEncoding(channels)
dec = transforms.MuLawExpanding(channels)
train_wav = enc(train_wav)
train_targets = enc(train_targets)
scaler = StandardScaler()
train_features = scaler.fit_transform(train_features)
train_wav = train_wav.reshape(-1, seq_M)
train_features = np.rollaxis(
train_features.reshape(-1, seq_M, features_size), 2, 1)
train_targets = train_targets.reshape(-1, seq_M)
train_wav = torch.from_numpy(train_wav).long()
train_features = torch.from_numpy(train_features).float()
train_targets = torch.from_numpy(train_targets).long()
