mindspore.dataset.audio.Flanger
- class mindspore.dataset.audio.Flanger(sample_rate, delay=0.0, depth=2.0, regen=0.0, width=71.0, speed=0.5, phase=25.0, modulation=Modulation.SINUSOIDAL, interpolation=Interpolation.LINEAR)[source]
Apply a flanger effect to the audio.
- Parameters
sample_rate (int) – Sampling rate of the waveform, e.g. 44100 (Hz).
delay (float, optional) – Desired delay in milliseconds (ms), range: [0, 30] (default=0.0).
depth (float, optional) – Desired delay depth in milliseconds (ms), range: [0, 10] (default=2.0).
regen (float, optional) – Desired regen (feedback gain) in dB, range: [-95, 95] (default=0.0).
width (float, optional) – Desired width (delay gain) in dB, range: [0, 100] (default=71.0).
speed (float, optional) – Modulation speed in Hz, range: [0.1, 10] (default=0.5).
phase (float, optional) – Percentage phase-shift for multi-channel, range: [0, 100] (default=25.0).
modulation (Modulation, optional) – Modulation of the input tensor (default=Modulation.SINUSOIDAL). It can be one of Modulation.SINUSOIDAL or Modulation.TRIANGULAR.
interpolation (Interpolation, optional) – Interpolation of the input tensor (default=Interpolation.LINEAR). It can be one of Interpolation.LINEAR or Interpolation.QUADRATIC.
Examples
>>> import numpy as np >>> >>> waveform = np.array([[2.716064453125e-03, 6.34765625e-03], [9.246826171875e-03, 1.0894775390625e-02]]) >>> numpy_slices_dataset = ds.NumpySlicesDataset(data=waveform, column_names=["audio"]) >>> transforms = [audio.Flanger(44100)] >>> numpy_slices_dataset = numpy_slices_dataset.map(operations=transforms, input_columns=["audio"])