transform__axismasking: Axis Masking

In torchaudio: R Interface to 'pytorch''s 'torchaudio'

Axis Masking

Description

Apply masking to a spectrogram.

Usage

transform__axismasking(mask_param, axis, iid_masks)

Arguments

mask_param	(int): Maximum possible length of the mask.
axis	(int): What dimension the mask is applied on.
iid_masks	(bool): Applies iid masks to each of the examples in the batch dimension. This option is applicable only when the input tensor is 4D.

Details

forward param: specgram (Tensor): Tensor of dimension (..., freq, time).

mask_value (float): Value to assign to the masked columns.

Value

Tensor: Masked spectrogram of dimensions (..., freq, time).

torchaudio documentation built on Feb. 16, 2023, 9:41 p.m.