audio_detokenizer/transformer/encoder.py

# Copyright (c) 2021 Mobvoi Inc (Binbin Zhang, Di Wu)
#               2022 Xingchen Song (sxc19@mails.tsinghua.edu.cn)
#               2024 Alibaba Inc (Xiang Lyu)
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#   http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
# Modified from ESPnet(https://github.com/espnet/espnet)
"""Encoder definition."""
from typing import Tuple

import torch
import torch.utils.checkpoint as ckpt

from .convolution import ConvolutionModule
from .encoder_layer import TransformerEncoderLayer
from .encoder_layer import ConformerEncoderLayer
from .positionwise_feed_forward import PositionwiseFeedForward

from ..utils.class_utils import (
    BAILING_EMB_CLASSES,
    BAILING_SUBSAMPLE_CLASSES,
    BAILING_ATTENTION_CLASSES,
    BAILING_ACTIVATION_CLASSES,
)
from ..utils.mask import make_pad_mask
from ..utils.mask import add_optional_chunk_mask

import torch.nn.functional as F
import logging

class KVCache(torch.nn.Module):
    def __init__(self, max_batch_size, n_heads, max_seq_short, max_seq_long, head_dim, dtype=torch.float32):
        super().__init__()
        cache_shape_short = (max_batch_size, n_heads, max_seq_short, head_dim)
        self.register_buffer('k_cache_short', torch.zeros(cache_shape_short, dtype=dtype, device='cuda'))
        self.register_buffer('v_cache_short', torch.zeros(cache_shape_short, dtype=dtype, device='cuda'))

        cache_shape_long = (max_batch_size, n_heads, max_seq_long, head_dim)
        self.register_buffer('k_cache_long', torch.zeros(cache_shape_long, dtype=dtype, device='cuda'))
        self.register_buffer('v_cache_long', torch.zeros(cache_shape_long, dtype=dtype, device='cuda'))

        self.max_seq_short = max_seq_short
        self.max_seq_long = max_seq_long

    def update(self, input_pos, k_val, v_val, is_infer_short):
        # input_pos: [S], k_val: [B, H, S, D]
        assert input_pos.shape[0] == k_val.shape[2]
        if is_infer_short:
            k_out = self.k_cache_short
            v_out = self.v_cache_short
        else:
            k_out = self.k_cache_long
            v_out = self.v_cache_long

        k_out[:, :, input_pos] = k_val
        v_out[:, :, input_pos] = v_val
        return k_out, v_out


class BaseEncoder(torch.nn.Module):

    def __init__(
        self,
        input_size: int,
        output_size: int = 256,
        attention_heads: int = 4,
        linear_units: int = 2048,
        num_blocks: int = 6,
        dropout_rate: float = 0.1,
        positional_dropout_rate: float = 0.1,
        attention_dropout_rate: float = 0.0,
        input_layer: str = "conv2d",
        pos_enc_layer_type: str = "abs_pos",
        normalize_before: bool = True,
        static_chunk_size: int = 0,
        use_dynamic_chunk: bool = False,
        global_cmvn: torch.nn.Module = None,
        use_dynamic_left_chunk: bool = False,
        gradient_checkpointing: bool = False,
        max_seq_short: int = 384,
        max_seq_long: int = 2048,
    ):
        """
        Args:
            input_size (int): input dim
            output_size (int): dimension of attention
            attention_heads (int): the number of heads of multi head attention
            linear_units (int): the hidden units number of position-wise feed
                forward
            num_blocks (int): the number of decoder blocks
            dropout_rate (float): dropout rate
            attention_dropout_rate (float): dropout rate in attention
            positional_dropout_rate (float): dropout rate after adding
                positional encoding
            input_layer (str): input layer type.
                optional [linear, conv2d, conv2d6, conv2d8]
            pos_enc_layer_type (str): Encoder positional encoding layer type.
                opitonal [abs_pos, scaled_abs_pos, rel_pos, no_pos]
            normalize_before (bool):
                True: use layer_norm before each sub-block of a layer.
                False: use layer_norm after each sub-block of a layer.
            static_chunk_size (int): chunk size for static chunk training and
                decoding
            use_dynamic_chunk (bool): whether use dynamic chunk size for
                training or not, You can only use fixed chunk(chunk_size > 0)
                or dyanmic chunk size(use_dynamic_chunk = True)
            global_cmvn (Optional[torch.nn.Module]): Optional GlobalCMVN module
            use_dynamic_left_chunk (bool): whether use dynamic left chunk in
                dynamic chunk training
            key_bias: whether use bias in attention.linear_k, False for whisper models.
            gradient_checkpointing: rerunning a forward-pass segment for each
                checkpointed segment during backward.
        """
        super().__init__()
        self._output_size = output_size

        self.global_cmvn = global_cmvn
        self.embed = BAILING_SUBSAMPLE_CLASSES[input_layer](
            input_size,
            output_size,
            dropout_rate,
            BAILING_EMB_CLASSES[pos_enc_layer_type](output_size,
                                                      positional_dropout_rate),
        )

        self.normalize_before = normalize_before
        self.after_norm = torch.nn.LayerNorm(output_size, eps=1e-5)
        self.static_chunk_size = static_chunk_size
        self.use_dynamic_chunk = use_dynamic_chunk
        self.use_dynamic_left_chunk = use_dynamic_left_chunk
        self.gradient_checkpointing = gradient_checkpointing
        self.attention_heads = attention_heads
        self.head_dim = output_size // attention_heads
        self.compiled_infer_short = None
        self.compiled_infer_long = None
        self.max_seq_short = max_seq_short
        self.max_seq_long = max_seq_long

    def setup_caches(self, max_seq_short, max_seq_long, dtype=torch.float32):
        # import pdb; pdb.set_trace()
        assert max_seq_short == self.max_seq_short and max_seq_long == self.max_seq_long
        for it in self.encoders:
            it.self_attn.kv_cache = KVCache(1, self.attention_heads, self.max_seq_short, self.max_seq_long,
                                            self.head_dim, dtype)

    def output_size(self) -> int:
        return self._output_size

    def forward(
        self,
        xs: torch.Tensor,
        xs_lens: torch.Tensor,
        decoding_chunk_size: int = 0,
        num_decoding_left_chunks: int = -1,
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """Embed positions in tensor.

        Args:
            xs: padded input tensor (B, T, D)
            xs_lens: input length (B)
            decoding_chunk_size: decoding chunk size for dynamic chunk
                0: default for training, use random dynamic chunk.
                <0: for decoding, use full chunk.
                >0: for decoding, use fixed chunk size as set.
            num_decoding_left_chunks: number of left chunks, this is for decoding,
            the chunk size is decoding_chunk_size.
                >=0: use num_decoding_left_chunks
                <0: use all left chunks
        Returns:
            encoder output tensor xs, and subsampled masks
            xs: padded output tensor (B, T' ~= T/subsample_rate, D)
            masks: torch.Tensor batch padding mask after subsample
                (B, 1, T' ~= T/subsample_rate)
        NOTE(xcsong):
            We pass the `__call__` method of the modules instead of `forward` to the
            checkpointing API because `__call__` attaches all the hooks of the module.
            https://discuss.pytorch.org/t/any-different-between-model-input-and-model-forward-input/3690/2
        """
        T = xs.size(1)
        masks = ~make_pad_mask(xs_lens, T).unsqueeze(1)  # (B, 1, T)
        if self.global_cmvn is not None:
            xs = self.global_cmvn(xs)
        xs, pos_emb, masks = self.embed(xs, masks)
        mask_pad = masks  # (B, 1, T/subsample_rate)
        chunk_masks = add_optional_chunk_mask(xs, masks,
                                              self.use_dynamic_chunk,
                                              self.use_dynamic_left_chunk,
                                              decoding_chunk_size,
                                              self.static_chunk_size,
                                              num_decoding_left_chunks)
        if self.gradient_checkpointing and self.training:
            xs = self.forward_layers_checkpointed(xs, chunk_masks, pos_emb,
                                                  mask_pad)
        else:
            xs = self.forward_layers(xs, chunk_masks, pos_emb, mask_pad)
        if self.normalize_before:
            xs = self.after_norm(xs)
        # Here we assume the mask is not changed in encoder layers, so just
        # return the masks before encoder layers, and the masks will be used
        # for cross attention with decoder later
        return xs, masks

    def forward_layers(self, xs: torch.Tensor, chunk_masks: torch.Tensor,
                       pos_emb: torch.Tensor,
                       mask_pad: torch.Tensor) -> torch.Tensor:
        for layer in self.encoders:
            xs, chunk_masks, _, _ = layer(xs, chunk_masks, pos_emb, mask_pad)
        return xs

    @torch.jit.ignore(drop=True)
    def forward_layers_checkpointed(self, xs: torch.Tensor,
                                    chunk_masks: torch.Tensor,
                                    pos_emb: torch.Tensor,
                                    mask_pad: torch.Tensor) -> torch.Tensor:
        for layer in self.encoders:
            xs, chunk_masks, _, _ = ckpt.checkpoint(layer.__call__, xs,
                                                    chunk_masks, pos_emb,
                                                    mask_pad)
        return xs

    def forward_chunk(
        self,
        xs: torch.Tensor,
        offset: int,
        required_cache_size: int,
        att_cache: torch.Tensor = torch.zeros(0, 0, 0, 0),
        cnn_cache: torch.Tensor = torch.zeros(0, 0, 0, 0),
        att_mask: torch.Tensor = torch.ones((0, 0, 0), dtype=torch.bool),
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        """ Forward just one chunk

        Args:
            xs (torch.Tensor): chunk input, with shape (b=1, time, mel-dim),
                where `time == (chunk_size - 1) * subsample_rate + \
                        subsample.right_context + 1`
            offset (int): current offset in encoder output time stamp
            required_cache_size (int): cache size required for next chunk
                compuation
                >=0: actual cache size
                <0: means all history cache is required
            att_cache (torch.Tensor): cache tensor for KEY & VALUE in
                transformer/conformer attention, with shape
                (elayers, head, cache_t1, d_k * 2), where
                `head * d_k == hidden-dim` and
                `cache_t1 == chunk_size * num_decoding_left_chunks`.
            cnn_cache (torch.Tensor): cache tensor for cnn_module in conformer,
                (elayers, b=1, hidden-dim, cache_t2), where
                `cache_t2 == cnn.lorder - 1`

        Returns:
            torch.Tensor: output of current input xs,
                with shape (b=1, chunk_size, hidden-dim).
            torch.Tensor: new attention cache required for next chunk, with
                dynamic shape (elayers, head, ?, d_k * 2)
                depending on required_cache_size.
            torch.Tensor: new conformer cnn cache required for next chunk, with
                same shape as the original cnn_cache.

        """
        assert xs.size(0) == 1
        # tmp_masks is just for interface compatibility
        tmp_masks = torch.ones(1,
                               xs.size(1),
                               device=xs.device,
                               dtype=torch.bool)
        tmp_masks = tmp_masks.unsqueeze(1)
        if self.global_cmvn is not None:
            xs = self.global_cmvn(xs)
        # NOTE(xcsong): Before embed, shape(xs) is (b=1, time, mel-dim)
        xs, pos_emb, _ = self.embed(xs, tmp_masks, offset)
        # NOTE(xcsong): After  embed, shape(xs) is (b=1, chunk_size, hidden-dim)
        elayers, cache_t1 = att_cache.size(0), att_cache.size(2)
        chunk_size = xs.size(1)
        attention_key_size = cache_t1 + chunk_size
        pos_emb = self.embed.position_encoding(offset=offset - cache_t1,
                                               size=attention_key_size)
        if required_cache_size < 0:
            next_cache_start = 0
        elif required_cache_size == 0:
            next_cache_start = attention_key_size
        else:
            next_cache_start = max(attention_key_size - required_cache_size, 0)
        r_att_cache = []
        r_cnn_cache = []
        for i, layer in enumerate(self.encoders):
            # NOTE(xcsong): Before layer.forward
            #   shape(att_cache[i:i + 1]) is (1, head, cache_t1, d_k * 2),
            #   shape(cnn_cache[i])       is (b=1, hidden-dim, cache_t2)
            xs, _, new_att_cache, new_cnn_cache = layer(
                xs,
                att_mask,
                pos_emb,
                att_cache=att_cache[i:i + 1] if elayers > 0 else att_cache,
                cnn_cache=cnn_cache[i] if cnn_cache.size(0) > 0 else cnn_cache)
            # NOTE(xcsong): After layer.forward
            #   shape(new_att_cache) is (1, head, attention_key_size, d_k * 2),
            #   shape(new_cnn_cache) is (b=1, hidden-dim, cache_t2)setup_caches
            r_att_cache.append(new_att_cache[:, :, next_cache_start:, :])
            r_cnn_cache.append(new_cnn_cache.unsqueeze(0))
        if self.normalize_before:
            xs = self.after_norm(xs)

        # NOTE(xcsong): shape(r_att_cache) is (elayers, head, ?, d_k * 2),
        #   ? may be larger than cache_t1, it depends on required_cache_size
        r_att_cache = torch.cat(r_att_cache, dim=0)
        # NOTE(xcsong): shape(r_cnn_cache) is (e, b=1, hidden-dim, cache_t2)
        r_cnn_cache = torch.cat(r_cnn_cache, dim=0)

        return (xs, r_att_cache, r_cnn_cache)

    def inference_layers(self, xs: torch.Tensor, chunk_masks: torch.Tensor,
                       pos_emb: torch.Tensor,
                       mask_pad: torch.Tensor) -> torch.Tensor:
        for layer in self.encoders:
            xs, chunk_masks, _ = layer.inference(xs, chunk_masks, pos_emb, mask_pad)
        return xs

    @torch.jit.ignore(drop=True)
    def inference_layers_checkpointed(self, xs: torch.Tensor,
                                    chunk_masks: torch.Tensor,
                                    pos_emb: torch.Tensor,
                                    mask_pad: torch.Tensor) -> torch.Tensor:
        for layer in self.encoders:
            xs, chunk_masks, _ = ckpt.checkpoint(layer.inference, xs,
                                                    chunk_masks, pos_emb,
                                                    mask_pad)
        return xs

    def inference_prefill(
        self,
        xs: torch.Tensor,
        offset: int,
        cache_offset: int,
        att_mask: torch.Tensor = torch.ones((0, 0, 0), dtype=torch.bool),
        fix_shape=False,
        is_infer_short=False,
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:
        assert xs.size(0) == 1
        # tmp_masks is just for interface compatibility
        tmp_masks = torch.ones(1,
                               xs.size(1),
                               device=xs.device,
                               dtype=torch.bool)
        tmp_masks = tmp_masks.unsqueeze(1)
        if self.global_cmvn is not None:
            xs = self.global_cmvn(xs)
        # NOTE(xcsong): Before embed, shape(xs) is (b=1, time, mel-dim)
        xs, pos_emb, _ = self.embed(xs, tmp_masks, offset)
        # NOTE(xcsong): After  embed, shape(xs) is (b=1, chunk_size, hidden-dim)
        chunk_size = xs.size(1)
        attention_key_size = cache_offset + chunk_size
        max_seq = self.max_seq_short if is_infer_short else self.max_seq_long
        if fix_shape:
            pos_emb = self.embed.position_encoding(offset=offset - cache_offset,
                                                size=attention_key_size)
            target_seq_len = max_seq * 2 - 1
            current_seq_len = pos_emb.size(1)
            padding_size = target_seq_len - current_seq_len
            pos_emb = F.pad(pos_emb, (0, 0, 0, padding_size))
        else:
            pos_emb = self.embed.position_encoding(offset=offset - cache_offset,
                                                size=attention_key_size)
        cache_offset = torch.arange(0, xs.shape[1], device=xs.device, dtype=torch.int)

        for i, layer in enumerate(self.encoders):
            xs, _, _ = layer.inference(
                xs,
                att_mask,
                pos_emb,
                cache_offset=cache_offset,
                is_infer_short=is_infer_short,
            )
        if self.normalize_before:
            xs = self.after_norm(xs)

        return xs

    def prepare_for_decode(
        self,
        xs: torch.Tensor,
        offset: int,
        cache_offset: int,
        is_infer_short: bool,
    ):
        # assert xs.size(0) == 1
        chunk_size = xs.size(1)
        attention_key_size = cache_offset + chunk_size
        max_seq = self.max_seq_short if is_infer_short else self.max_seq_long
        # tmp_masks = torch.ones(1, chunk_size,
        #                     device=xs.device, dtype=torch.bool)
        # tmp_masks = tmp_masks.unsqueeze(1)
        # if self.global_cmvn is not None:
        #     xs = self.global_cmvn(xs)
        # xs, _, _ = self.embed(xs, tmp_masks, offset)
        xs = self.embed.out(xs) * self.embed.pos_enc.xscale
        pos_emb = self.embed.fix_position_encoding(offset=offset - cache_offset,
                                                   size=attention_key_size, max_len=max_seq)
        # print("xs, pos_emb", xs.device, pos_emb.device)
        return xs, pos_emb

    def step_infer_short(
        self,
        xs: torch.Tensor,
        pos_emb: torch.Tensor,
        cache_offset: torch.Tensor,
        att_mask: torch.Tensor = torch.ones((0, 0, 0), dtype=torch.bool),
    ):
        for i, layer in enumerate(self.encoders):
            xs, _, _ = layer.inference(
                xs,
                att_mask,
                pos_emb,
                cache_offset=cache_offset,
                is_infer_short = True,
            )
        if self.normalize_before:
            xs = self.after_norm(xs)
        return xs

    def step_infer_long(
        self,
        xs: torch.Tensor,
        pos_emb: torch.Tensor,
        cache_offset: torch.Tensor,
        att_mask: torch.Tensor = torch.ones((0, 0, 0), dtype=torch.bool),
    ):
        for i, layer in enumerate(self.encoders):
            xs, _, _ = layer.inference(
                xs,
                att_mask,
                pos_emb,
                cache_offset=cache_offset,
                is_infer_short=False,
            )
        if self.normalize_before:
            xs = self.after_norm(xs)
        return xs

    def inference_decode_step(
        self,
        xs: torch.Tensor,
        offset: int,
        cache_offset: int,
        att_mask,
        is_infer_short,
    ) -> Tuple[torch.Tensor, torch.Tensor, torch.Tensor]:

        xs, pos_emb = self.prepare_for_decode(xs, offset, cache_offset, is_infer_short)
        cache_offset = torch.tensor([cache_offset], device=xs.device, dtype=torch.int32)

        # print("xs, att_mask: ", xs.shape, att_mask.shape)

        if is_infer_short:
            if self.compiled_infer_short == None:
                self.compiled_infer_short = torch.compile(self.step_infer_short, mode="reduce-overhead", fullgraph=True)
            ret = self.compiled_infer_short(xs, pos_emb, cache_offset, att_mask)
        elif not is_infer_short:
            if self.compiled_infer_long == None:
                self.compiled_infer_long = torch.compile(self.step_infer_long, mode="reduce-overhead", fullgraph=True)
            ret = self.compiled_infer_long(xs, pos_emb, cache_offset, att_mask)

        return ret.clone()

    def forward_chunk_by_chunk(
        self,
        xs: torch.Tensor,
        decoding_chunk_size: int,
        num_decoding_left_chunks: int = -1,
    ) -> Tuple[torch.Tensor, torch.Tensor]:
        """ Forward input chunk by chunk with chunk_size like a streaming
            fashion

        Here we should pay special attention to computation cache in the
        streaming style forward chunk by chunk. Three things should be taken
        into account for computation in the current network:
            1. transformer/conformer encoder layers output cache
            2. convolution in conformer
            3. convolution in subsampling

        However, we don't implement subsampling cache for:
            1. We can control subsampling module to output the right result by
               overlapping input instead of cache left context, even though it
               wastes some computation, but subsampling only takes a very
               small fraction of computation in the whole model.
            2. Typically, there are several covolution layers with subsampling
               in subsampling module, it is tricky and complicated to do cache
               with different convolution layers with different subsampling
               rate.
            3. Currently, nn.Sequential is used to stack all the convolution
               layers in subsampling, we need to rewrite it to make it work
               with cache, which is not prefered.
        Args:
            xs (torch.Tensor): (1, max_len, dim)
            chunk_size (int): decoding chunk size
        """
        assert decoding_chunk_size > 0
        # The model is trained by static or dynamic chunk
        assert self.static_chunk_size > 0 or self.use_dynamic_chunk
        subsampling = self.embed.subsampling_rate
        context = self.embed.right_context + 1  # Add current frame
        stride = subsampling * decoding_chunk_size
        decoding_window = (decoding_chunk_size - 1) * subsampling + context
        num_frames = xs.size(1)
        att_cache: torch.Tensor = torch.zeros((0, 0, 0, 0), device=xs.device)
        cnn_cache: torch.Tensor = torch.zeros((0, 0, 0, 0), device=xs.device)
        outputs = []
        offset = 0
        required_cache_size = decoding_chunk_size * num_decoding_left_chunks

        # Feed forward overlap input step by step
        for cur in range(0, num_frames - context + 1, stride):
            end = min(cur + decoding_window, num_frames)
            chunk_xs = xs[:, cur:end, :]
            (y, att_cache,
             cnn_cache) = self.forward_chunk(chunk_xs, offset,
                                             required_cache_size, att_cache,
                                             cnn_cache)
            outputs.append(y)
            offset += y.size(1)
        ys = torch.cat(outputs, 1)
        masks = torch.ones((1, 1, ys.size(1)),
                           device=ys.device,
                           dtype=torch.bool)
        return ys, masks


class TransformerEncoder(BaseEncoder):
    """Transformer encoder module."""

    def __init__(
        self,
        input_size: int,
        output_size: int = 256,
        attention_heads: int = 4,
        linear_units: int = 2048,
        num_blocks: int = 6,
        dropout_rate: float = 0.1,
        positional_dropout_rate: float = 0.1,
        attention_dropout_rate: float = 0.0,
        input_layer: str = "conv2d",
        pos_enc_layer_type: str = "abs_pos",
        normalize_before: bool = True,
        static_chunk_size: int = 0,
        use_dynamic_chunk: bool = False,
        global_cmvn: torch.nn.Module = None,
        use_dynamic_left_chunk: bool = False,
        key_bias: bool = True,
        selfattention_layer_type: str = "selfattn",
        activation_type: str = "relu",
        gradient_checkpointing: bool = False,
        max_seq_short: int = 384,
        max_seq_long: int = 2048,
    ):
        """ Construct TransformerEncoder

        See Encoder for the meaning of each parameter.
        """
        super().__init__(input_size, output_size, attention_heads,
                         linear_units, num_blocks, dropout_rate,
                         positional_dropout_rate, attention_dropout_rate,
                         input_layer, pos_enc_layer_type, normalize_before,
                         static_chunk_size, use_dynamic_chunk, global_cmvn,
                         use_dynamic_left_chunk, gradient_checkpointing, max_seq_short, max_seq_long)
        activation = BAILING_ACTIVATION_CLASSES[activation_type]()
        self.encoders = torch.nn.ModuleList([
            TransformerEncoderLayer(
                output_size,
                BAILING_ATTENTION_CLASSES[selfattention_layer_type](attention_heads,
                                                                      output_size,
                                                                      attention_dropout_rate,
                                                                      key_bias),
                PositionwiseFeedForward(output_size, linear_units,
                                        dropout_rate, activation),
                dropout_rate, normalize_before) for _ in range(num_blocks)
        ])


class ConformerEncoder(BaseEncoder):
    """Conformer encoder module."""

    def __init__(
        self,
        input_size: int,
        output_size: int = 256,
        attention_heads: int = 4,
        linear_units: int = 2048,
        num_blocks: int = 6,
        dropout_rate: float = 0.1,
        positional_dropout_rate: float = 0.1,
        attention_dropout_rate: float = 0.0,
        input_layer: str = "conv2d",
        pos_enc_layer_type: str = "rel_pos",
        normalize_before: bool = True,
        static_chunk_size: int = 0,
        use_dynamic_chunk: bool = False,
        global_cmvn: torch.nn.Module = None,
        use_dynamic_left_chunk: bool = False,
        positionwise_conv_kernel_size: int = 1,
        macaron_style: bool = True,
        selfattention_layer_type: str = "rel_selfattn",
        activation_type: str = "swish",
        use_cnn_module: bool = True,
        cnn_module_kernel: int = 15,
        causal: bool = False,
        cnn_module_norm: str = "batch_norm",
        key_bias: bool = True,
        gradient_checkpointing: bool = False,
        max_seq_short: int = 384,
        max_seq_long: int = 2048,
    ):
        """Construct ConformerEncoder

        Args:
            input_size to use_dynamic_chunk, see in BaseEncoder
            positionwise_conv_kernel_size (int): Kernel size of positionwise
                conv1d layer.
            macaron_style (bool): Whether to use macaron style for
                positionwise layer.
            selfattention_layer_type (str): Encoder attention layer type,
                the parameter has no effect now, it's just for configure
                compatibility.
            activation_type (str): Encoder activation function type.
            use_cnn_module (bool): Whether to use convolution module.
            cnn_module_kernel (int): Kernel size of convolution module.
            causal (bool): whether to use causal convolution or not.
            key_bias: whether use bias in attention.linear_k, False for whisper models.
        """
        super().__init__(input_size, output_size, attention_heads,
                         linear_units, num_blocks, dropout_rate,
                         positional_dropout_rate, attention_dropout_rate,
                         input_layer, pos_enc_layer_type, normalize_before,
                         static_chunk_size, use_dynamic_chunk, global_cmvn,
                         use_dynamic_left_chunk, gradient_checkpointing, max_seq_short, max_seq_long)
        activation = BAILING_ACTIVATION_CLASSES[activation_type]()

        # self-attention module definition
        encoder_selfattn_layer_args = (
            attention_heads,
            output_size,
            attention_dropout_rate,
            key_bias,
        )
        # feed-forward module definition
        positionwise_layer_args = (
            output_size,
            linear_units,
            dropout_rate,
            activation,
        )
        # convolution module definition
        convolution_layer_args = (output_size, cnn_module_kernel, activation,
                                  cnn_module_norm, causal)

        self.encoders = torch.nn.ModuleList([
            ConformerEncoderLayer(
                output_size,
                BAILING_ATTENTION_CLASSES[selfattention_layer_type](
                    *encoder_selfattn_layer_args),
                PositionwiseFeedForward(*positionwise_layer_args),
                PositionwiseFeedForward(
                    *positionwise_layer_args) if macaron_style else None,
                ConvolutionModule(
                    *convolution_layer_args) if use_cnn_module else None,
                dropout_rate,
                normalize_before,
            ) for _ in range(num_blocks)
        ])