o *iT'@sddlmZmZddlZdgZdejdejfddZGdd d ejjZ Gd d d ejjZ Gd d d ejjZ GdddejjZ dS))OptionalTupleN ConformerlengthsreturncCsF|jd}tt|}tj||j|jd||| dk}|S)Nr)devicedtype) shapeinttorchmaxitemZarangerrexpandZ unsqueeze)rZ batch_size max_lengthZ padding_maskrj/var/www/html/eduruby.in/lip-sync/lip-sync-env/lib/python3.10/site-packages/torchaudio/models/conformer.py_lengths_to_padding_mask s rcs\eZdZdZ   ddededededed ed d ffd d Zdej d ej fddZ Z S)_ConvolutionModuleaConformer convolution module. Args: input_dim (int): input dimension. num_channels (int): number of depthwise convolution layer input channels. depthwise_kernel_size (int): kernel size of depthwise convolution layer. dropout (float, optional): dropout probability. (Default: 0.0) bias (bool, optional): indicates whether to add bias term to each convolution layer. (Default: ``False``) use_group_norm (bool, optional): use GroupNorm rather than BatchNorm. (Default: ``False``) F input_dim num_channelsdepthwise_kernel_sizedropoutbiasuse_group_normrNcst|dddkrtdtj||_tjtjj|d|ddd|dtjj ddtjj|||d|dd||d|rHtjj d|dntj |tj tjj||ddd|d tj ||_dS) Nr rz