Differences with torchaudio.transforms.InverseMelScale

torchaudio.transforms.InverseMelScale

class torchaudio.transforms.InverseMelScale(n_stft: int, n_mels: int = 128, sample_rate: int = 16000, f_min: float = 0.0, f_max: Optional[float] = None,
                                            max_iter: int = 100000, tolerance_loss: float = 1e-05, tolerance_change: float = 1e-08, sgdargs: Optional[dict] = None,
                                            norm: Optional[str] = None)

For more information, see torchaudio.transforms.InverseMelScale.

mindspore.dataset.audio.InverseMelScale

class mindspore.dataset.audio.InverseMelScale(n_stft, n_mels=128, sample_rate=16000, f_min=0.0, f_max=None,
                                              max_iter=100000, tolerance_loss=1e-5, tolerance_change=1e-8, sgdargs=None,
                                              norm=NormType.NONE, mel_type=MelType.HTK)

For more information, see mindspore.dataset.audio.InverseMelScale.

Differences

PyTorch: Solve for a normal STFT from a mel frequency STFT, using a conversion matrix.

MindSpore: Solve for a normal STFT from a mel frequency STFT, using a conversion matrix. Mel scale can be specified.

Categories	Subcategories	PyTorch	MindSpore	Difference
Parameter	Parameter1	n_stft	n_stft	-
	Parameter2	n_mels	n_mels	-
	Parameter3	sample_rate	sample_rate	-
	Parameter4	f_min	f_min	-
	Parameter5	f_max	f_max	-
	Parameter6	max_iter	max_iter	-
	Parameter7	tolerance_loss	tolerance_loss	-
	Parameter8	tolerance_change	tolerance_change	-
	Parameter9	sgdargs	sgdargs	-
	Parameter10	norm	norm	-
	Parameter11	-	mel_type	Mel scale to use

Code Example

import numpy as np

fake_input = np.array([[1., 1.],
                       [0., 0.],
                       [1., 1.],
                       [1., 1.]]).astype(np.float32)

# PyTorch
import torch
import torchaudio.transforms as T
torch.manual_seed(1)

transformer = T.InverseMelScale(n_stft=2, n_mels=4)
torch_result = transformer(torch.from_numpy(fake_input))
print(torch_result)
# Out: tensor([[0.7576, 0.4031],
#              [0.2793, 0.7347]])

# MindSpore
import mindspore as ms
import mindspore.dataset.audio as audio
ms.dataset.config.set_seed(3)

transformer = audio.InverseMelScale(n_stft=2, n_mels=4)
ms_result = transformer(fake_input)
print(ms_result)
# Out: [[[0.5507979  0.07072488]
#        [0.7081478  0.8399491 ]]]