mindspore.dataset.audio.MuLawEncoding

class mindspore.dataset.audio.MuLawEncoding(quantization_channels=256)[source]

Encode signal based on mu-law companding.

Parameters: quantization_channels (int, optional) – Number of channels, which must be positive. Default: 256.

Examples

>>> import numpy as np
>>>
>>> waveform = np.random.random([1, 3, 4])
>>> numpy_slices_dataset = ds.NumpySlicesDataset(data=waveform, column_names=["audio"])
>>> transforms = [audio.MuLawEncoding()]
>>> numpy_slices_dataset = numpy_slices_dataset.map(operations=transforms, input_columns=["audio"])