modules/wenet_extractor/efficient_conformer/subsampling.py

# This module is from [WeNet](https://github.com/wenet-e2e/wenet).

# ## Citations

# ```bibtex
# @inproceedings{yao2021wenet,
#   title={WeNet: Production oriented Streaming and Non-streaming End-to-End Speech Recognition Toolkit},
#   author={Yao, Zhuoyuan and Wu, Di and Wang, Xiong and Zhang, Binbin and Yu, Fan and Yang, Chao and Peng, Zhendong and Chen, Xiaoyu and Xie, Lei and Lei, Xin},
#   booktitle={Proc. Interspeech},
#   year={2021},
#   address={Brno, Czech Republic },
#   organization={IEEE}
# }

# @article{zhang2022wenet,
#   title={WeNet 2.0: More Productive End-to-End Speech Recognition Toolkit},
#   author={Zhang, Binbin and Wu, Di and Peng, Zhendong and Song, Xingchen and Yao, Zhuoyuan and Lv, Hang and Xie, Lei and Yang, Chao and Pan, Fuping and Niu, Jianwei},
#   journal={arXiv preprint arXiv:2203.15455},
#   year={2022}
# }
#


"""Subsampling layer definition."""

from typing import Tuple, Union

import torch
from modules.wenet_extractor.transformer.subsampling import BaseSubsampling


class Conv2dSubsampling2(BaseSubsampling):
    """Convolutional 2D subsampling (to 1/4 length).

    Args:
        idim (int): Input dimension.
        odim (int): Output dimension.
        dropout_rate (float): Dropout rate.

    """

    def __init__(
        self, idim: int, odim: int, dropout_rate: float, pos_enc_class: torch.nn.Module
    ):
        """Construct an Conv2dSubsampling4 object."""
        super().__init__()
        self.conv = torch.nn.Sequential(torch.nn.Conv2d(1, odim, 3, 2), torch.nn.ReLU())
        self.out = torch.nn.Sequential(torch.nn.Linear(odim * ((idim - 1) // 2), odim))
        self.pos_enc = pos_enc_class
        # The right context for every conv layer is computed by:
        # (kernel_size - 1) * frame_rate_of_this_layer
        self.subsampling_rate = 2
        # 2 = (3 - 1) * 1
        self.right_context = 2

    def forward(
        self,
        x: torch.Tensor,
        x_mask: torch.Tensor,
        offset: Union[int, torch.Tensor] = 0,
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        """Subsample x.

        Args:
            x (torch.Tensor): Input tensor (#batch, time, idim).
            x_mask (torch.Tensor): Input mask (#batch, 1, time).

        Returns:
            torch.Tensor: Subsampled tensor (#batch, time', odim),
                where time' = time // 2.
            torch.Tensor: Subsampled mask (#batch, 1, time'),
                where time' = time // 2.
            torch.Tensor: positional encoding

        """
        x = x.unsqueeze(1)  # (b, c=1, t, f)
        x = self.conv(x)
        b, c, t, f = x.size()
        x = self.out(x.transpose(1, 2).contiguous().view(b, t, c * f))
        x, pos_emb = self.pos_enc(x, offset)
        return x, pos_emb, x_mask[:, :, :-2:2]