This function implements 1d convolution. Arguments --------- out_channels : int It is the number of output channels. kernel_s

An implementation of TDNN. Arguments ---------- in_channels : int Number of input channels. out_channels : int The nu

Implementation of a sine-based periodic activation function Shape: - Input: (B, C, T) - Output: (B, C, T), same shape as the

A modified Snake function which uses separate parameters for the magnitude of the periodic components Shape: - Input: (B, C, T)

Applies 1d batch normalization to the input tensor. Arguments --------- input_shape : tuple The expected shape of the input. Alte

Encoder layer module. Args: size (int): Input dimension. self_attn (torch.nn.Module): Self-attention module instance.

An implementation of the speaker embedding model in a paper. "ECAPA-TDNN: Emphasized Channel Attention, Propagation and Aggregation in TDNN Ba

Relative positional encoding module (new implementation). Details can be found in https://github.com/espnet/espnet/pull/2816. See : Appendix

Linear transform the input without subsampling Args: idim (int): Input dimension. odim (int): Output dimension. dropout_ra

Positionwise feed forward layer. FeedForward are appied on each position of the sequence. The output dim is same with the input dim. Arg

This class implements an attentive statistic pooling layer for each channel. It returns the concatenated mean and std of the input tensor. Ar

Diffusion model with a Transformer backbone.

A DiT block with adaptive layer norm zero (adaLN-Zero) conditioning.

The final layer of DiT.

Override GPT2LMHeadModel to allow for prefix conditioning.

Override GPT2LMHeadModel to allow for prefix conditioning.

An implementation of Res2NetBlock w/ dilation. Arguments --------- in_channels : int The number of channels expected in the input

An implementation of squeeze-and-excitation block. Arguments --------- in_channels : int The number of input channels. se_cha

An implementation of building block in ECAPA-TDNN, i.e., TDNN-Res2Net-TDNN-SEBlock. Arguments ---------- out_channels: int Th

Embeds scalar timesteps into vector representations.

A 1D upsampling layer with an optional convolution. Parameters: channels (`int`): number of channels in the inputs and output

Computes a linear transformation y = wx + b. Arguments --------- n_neurons : int It is the number of output neurons (i.e, the dim