mindspore.dataset.audio.Dither
- class mindspore.dataset.audio.Dither(density_function=DensityFunction.TPDF, noise_shaping=False)[source]
Dither increases the perceived dynamic range of audio stored at a particular bit-depth by eliminating nonlinear truncation distortion.
- Parameters
density_function (DensityFunction, optional) – The density function of a continuous random variable, can be
DensityFunction.TPDF
(Triangular Probability Density Function),DensityFunction.RPDF
(Rectangular Probability Density Function) orDensityFunction.GPDF
(Gaussian Probability Density Function). Default:DensityFunction.TPDF
.noise_shaping (bool, optional) – A filtering process that shapes the spectral energy of quantisation error. Default:
False
.
- Raises
TypeError – If density_function is not of type
mindspore.dataset.audio.DensityFunction
.TypeError – If noise_shaping is not of type bool.
RuntimeError – If input tensor is not in shape of <…, time>.
- Supported Platforms:
CPU
Examples
>>> import numpy as np >>> import mindspore.dataset as ds >>> import mindspore.dataset.audio as audio >>> >>> waveform = np.array([[1, 2, 3], [4, 5, 6]]) >>> numpy_slices_dataset = ds.NumpySlicesDataset(data=waveform, column_names=["audio"]) >>> transforms = [audio.Dither()] >>> numpy_slices_dataset = numpy_slices_dataset.map(operations=transforms, input_columns=["audio"])
- Tutorial Examples: