[ASR]add squeezeformer model (#2755)

* add squeezeformer model

* change CodeStyle, test=asr

* change CodeStyle, test=asr

* fix subsample rate error, test=asr

* merge classes as required, test=asr

* change CodeStyle, test=asr

* fix missing code, test=asr

* split code to new file, test=asr

* remove rel_shift, test=asr

examples/aishell/asr1/conf/chunk_squeezeformer.yaml

@@ -0,0 +1,98 @@
+############################################
+#           Network Architecture           #
+############################################
+cmvn_file: 
+cmvn_file_type: "json"
+# encoder related
+encoder: squeezeformer
+encoder_conf:
+    encoder_dim: 256    # dimension of attention
+    output_size: 256    # dimension of output
+    attention_heads: 4
+    num_blocks: 12      # the number of encoder blocks
+    reduce_idx: 5
+    recover_idx: 11
+    feed_forward_expansion_factor: 8
+    input_dropout_rate: 0.1
+    feed_forward_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    adaptive_scale: true
+    cnn_module_kernel: 31
+    normalize_before: false
+    activation_type: 'swish'
+    pos_enc_layer_type: 'rel_pos'
+    time_reduction_layer_type: 'stream'
+    causal: true
+    use_dynamic_chunk: true
+    use_dynamic_left_chunk: false
+
+# decoder related
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1  # sublayer output dropout
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1     # label smoothing option
+    length_normalized_loss: false
+    init_type: 'kaiming_uniform' # !Warning: need to convergence
+
+###########################################
+#                   Data                  #
+###########################################
+
+train_manifest: data/manifest.train
+dev_manifest: data/manifest.dev
+test_manifest: data/manifest.test
+
+
+###########################################
+#              Dataloader                 #
+###########################################
+
+vocab_filepath: data/lang_char/vocab.txt 
+spm_model_prefix: ''
+unit_type: 'char'
+preprocess_config: conf/preprocess.yaml
+feat_dim: 80
+stride_ms: 10.0
+window_ms: 25.0
+sortagrad: 0 # Feed samples from shortest to longest ; -1: enabled for all epochs, 0: disabled, other: enabled for 'other' epochs 
+batch_size: 32
+maxlen_in: 512  # if input length  > maxlen-in, batchsize is automatically reduced
+maxlen_out: 150  # if output length > maxlen-out, batchsize is automatically reduced
+minibatches: 0 # for debug
+batch_count: auto
+batch_bins: 0
+batch_frames_in: 0
+batch_frames_out: 0
+batch_frames_inout: 0
+num_workers: 2
+subsampling_factor: 1
+num_encs: 1
+
+###########################################
+#                 Training                #
+###########################################
+n_epoch: 240 
+accum_grad: 1
+global_grad_clip: 5.0
+dist_sampler: True
+optim: adam
+optim_conf:
+  lr: 0.001
+  weight_decay: 1.0e-6
+scheduler: warmuplr
+scheduler_conf:
+  warmup_steps: 25000
+  lr_decay: 1.0
+log_interval: 100
+checkpoint:
+  kbest_n: 50
+  latest_n: 5

examples/aishell/asr1/conf/squeezeformer.yaml

@@ -0,0 +1,93 @@
+############################################
+#           Network Architecture           #
+############################################
+cmvn_file: 
+cmvn_file_type: "json"
+# encoder related
+encoder: squeezeformer
+encoder_conf:
+    encoder_dim: 256    # dimension of attention
+    output_size: 256    # dimension of output
+    attention_heads: 4
+    num_blocks: 12      # the number of encoder blocks
+    reduce_idx: 5
+    recover_idx: 11
+    feed_forward_expansion_factor: 8
+    input_dropout_rate: 0.1
+    feed_forward_dropout_rate: 0.1
+    attention_dropout_rate: 0.1
+    adaptive_scale: true
+    cnn_module_kernel: 31
+    normalize_before: false
+    activation_type: 'swish'
+    pos_enc_layer_type: 'rel_pos'
+    time_reduction_layer_type: 'conv1d'
+
+# decoder related
+decoder: transformer
+decoder_conf:
+    attention_heads: 4
+    linear_units: 2048
+    num_blocks: 6
+    dropout_rate: 0.1
+    positional_dropout_rate: 0.1
+    self_attention_dropout_rate: 0.0
+    src_attention_dropout_rate: 0.0
+
+# hybrid CTC/attention
+model_conf:
+    ctc_weight: 0.3
+    lsm_weight: 0.1     # label smoothing option
+    length_normalized_loss: false
+    init_type: 'kaiming_uniform' # !Warning: need to convergence
+
+###########################################
+#                   Data                  #
+###########################################
+train_manifest: data/manifest.train
+dev_manifest: data/manifest.dev
+test_manifest: data/manifest.test
+
+###########################################
+#              Dataloader                 #
+###########################################
+vocab_filepath: data/lang_char/vocab.txt 
+spm_model_prefix: ''
+unit_type: 'char'
+preprocess_config: conf/preprocess.yaml
+feat_dim: 80
+stride_ms: 10.0
+window_ms: 25.0
+sortagrad: 0 # Feed samples from shortest to longest ; -1: enabled for all epochs, 0: disabled, other: enabled for 'other' epochs 
+batch_size: 32
+maxlen_in: 512  # if input length  > maxlen-in, batchsize is automatically reduced
+maxlen_out: 150  # if output length > maxlen-out, batchsize is automatically reduced
+minibatches: 0 # for debug
+batch_count: auto
+batch_bins: 0 
+batch_frames_in: 0
+batch_frames_out: 0
+batch_frames_inout: 0
+num_workers: 2
+subsampling_factor: 1
+num_encs: 1
+
+###########################################
+#                Training                 #
+###########################################
+n_epoch: 150 
+accum_grad: 8
+global_grad_clip: 5.0
+dist_sampler: False
+optim: adam
+optim_conf:
+  lr: 0.002
+  weight_decay: 1.0e-6
+scheduler: warmuplr
+scheduler_conf:
+  warmup_steps: 25000
+  lr_decay: 1.0
+log_interval: 100
+checkpoint:
+  kbest_n: 50
+  latest_n: 5

paddlespeech/s2t/models/u2/u2.py

@@ -43,6 +43,7 @@ from paddlespeech.s2t.modules.ctc import CTCDecoderBase
 from paddlespeech.s2t.modules.decoder import BiTransformerDecoder
 from paddlespeech.s2t.modules.decoder import TransformerDecoder
 from paddlespeech.s2t.modules.encoder import ConformerEncoder
+from paddlespeech.s2t.modules.encoder import SqueezeformerEncoder
 from paddlespeech.s2t.modules.encoder import TransformerEncoder
 from paddlespeech.s2t.modules.initializer import DefaultInitializerContext
 from paddlespeech.s2t.modules.loss import LabelSmoothingLoss
@@ -905,6 +906,9 @@ class U2Model(U2DecodeModel):
         elif encoder_type == 'conformer':
             encoder = ConformerEncoder(
                 input_dim, global_cmvn=global_cmvn, **configs['encoder_conf'])
+        elif encoder_type == 'squeezeformer':
+            encoder = SqueezeformerEncoder(
+                input_dim, global_cmvn=global_cmvn, **configs['encoder_conf'])
         else:
             raise ValueError(f"not support encoder type:{encoder_type}")
 

paddlespeech/s2t/modules/attention.py

@@ -200,7 +200,12 @@ class MultiHeadedAttention(nn.Layer):
 class RelPositionMultiHeadedAttention(MultiHeadedAttention):
     """Multi-Head Attention layer with relative position encoding."""
 
-    def __init__(self, n_head, n_feat, dropout_rate):
+    def __init__(self,
+                 n_head,
+                 n_feat,
+                 dropout_rate,
+                 adaptive_scale=False,
+                 init_weights=False):
         """Construct an RelPositionMultiHeadedAttention object.
         Paper: https://arxiv.org/abs/1901.02860
         Args:
@@ -223,6 +228,39 @@ class RelPositionMultiHeadedAttention(MultiHeadedAttention):
         pos_bias_v = self.create_parameter(
             (self.h, self.d_k), default_initializer=I.XavierUniform())
         self.add_parameter('pos_bias_v', pos_bias_v)
+        self.adaptive_scale = adaptive_scale
+        if self.adaptive_scale:
+            ada_scale = self.create_parameter(
+                [1, 1, n_feat], default_initializer=I.Constant(1.0))
+            self.add_parameter('ada_scale', ada_scale)
+            ada_bias = self.create_parameter(
+                [1, 1, n_feat], default_initializer=I.Constant(0.0))
+            self.add_parameter('ada_bias', ada_bias)
+        if init_weights:
+            self.init_weights()
+
+    def init_weights(self):
+        input_max = (self.h * self.d_k)**-0.5
+        self.linear_q._param_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_q._bias_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_k._param_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_k._bias_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_v._param_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_v._bias_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_pos._param_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_pos._bias_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_out._param_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
+        self.linear_out._bias_attr = paddle.nn.initializer.Uniform(
+            low=-input_max, high=input_max)
 
     def rel_shift(self, x, zero_triu: bool=False):
         """Compute relative positinal encoding.
@@ -273,6 +311,11 @@ class RelPositionMultiHeadedAttention(MultiHeadedAttention):
                 where `cache_t == chunk_size * num_decoding_left_chunks`
                 and `head * d_k == size`
         """
+        if self.adaptive_scale:
+            query = self.ada_scale * query + self.ada_bias
+            key = self.ada_scale * key + self.ada_bias
+            value = self.ada_scale * value + self.ada_bias
+
         q, k, v = self.forward_qkv(query, key, value)
         # q = q.transpose([0, 2, 1, 3])  # (batch, time1, head, d_k)
 

paddlespeech/s2t/modules/conformer_convolution.py

@@ -18,6 +18,7 @@ from typing import Tuple
 
 import paddle
 from paddle import nn
+from paddle.nn import initializer as I
 from typeguard import check_argument_types
 
 from paddlespeech.s2t.modules.align import BatchNorm1D
@@ -39,7 +40,9 @@ class ConvolutionModule(nn.Layer):
                  activation: nn.Layer=nn.ReLU(),
                  norm: str="batch_norm",
                  causal: bool=False,
-                 bias: bool=True):
+                 bias: bool=True,
+                 adaptive_scale: bool=False,
+                 init_weights: bool=False):
         """Construct an ConvolutionModule object.
         Args:
             channels (int): The number of channels of conv layers.
@@ -51,6 +54,18 @@ class ConvolutionModule(nn.Layer):
         """
         assert check_argument_types()
         super().__init__()
+        self.bias = bias
+        self.channels = channels
+        self.kernel_size = kernel_size
+        self.adaptive_scale = adaptive_scale
+        if self.adaptive_scale:
+            ada_scale = self.create_parameter(
+                [1, 1, channels], default_initializer=I.Constant(1.0))
+            self.add_parameter('ada_scale', ada_scale)
+            ada_bias = self.create_parameter(
+                [1, 1, channels], default_initializer=I.Constant(0.0))
+            self.add_parameter('ada_bias', ada_bias)
+
         self.pointwise_conv1 = Conv1D(
             channels,
             2 * channels,
@@ -105,6 +120,28 @@ class ConvolutionModule(nn.Layer):
         )
         self.activation = activation
 
+        if init_weights:
+            self.init_weights()
+
+    def init_weights(self):
+        pw_max = self.channels**-0.5
+        dw_max = self.kernel_size**-0.5
+        self.pointwise_conv1._param_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+        if self.bias:
+            self.pointwise_conv1._bias_attr = paddle.nn.initializer.Uniform(
+                low=-pw_max, high=pw_max)
+        self.depthwise_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        if self.bias:
+            self.depthwise_conv._bias_attr = paddle.nn.initializer.Uniform(
+                low=-dw_max, high=dw_max)
+        self.pointwise_conv2._param_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+        if self.bias:
+            self.pointwise_conv2._bias_attr = paddle.nn.initializer.Uniform(
+                low=-pw_max, high=pw_max)
+
     def forward(
             self,
             x: paddle.Tensor,
@@ -123,6 +160,9 @@ class ConvolutionModule(nn.Layer):
             paddle.Tensor: Output tensor (#batch, time, channels).
             paddle.Tensor: Output cache tensor (#batch, channels, time')
         """
+        if self.adaptive_scale:
+            x = self.ada_scale * x + self.ada_bias
+
         # exchange the temporal dimension and the feature dimension
         x = x.transpose([0, 2, 1])  # [B, C, T]
 

paddlespeech/s2t/modules/conv2d.py

@@ -0,0 +1,62 @@
+from typing import Optional
+from typing import Union
+
+import paddle
+import paddle.nn.functional as F
+from paddle.nn.layer.conv import _ConvNd
+
+__all__ = ['Conv2DValid']
+
+
+class Conv2DValid(_ConvNd):
+    """
+    Conv2d operator for VALID mode padding.
+    """
+
+    def __init__(self,
+                 in_channels: int,
+                 out_channels: int,
+                 kernel_size: int,
+                 stride: int=1,
+                 padding: Union[str, int]=0,
+                 dilation: int=1,
+                 groups: int=1,
+                 padding_mode: str='zeros',
+                 weight_attr=None,
+                 bias_attr=None,
+                 data_format="NCHW",
+                 valid_trigx: bool=False,
+                 valid_trigy: bool=False) -> None:
+        super(Conv2DValid, self).__init__(
+            in_channels,
+            out_channels,
+            kernel_size,
+            False,
+            2,
+            stride=stride,
+            padding=padding,
+            padding_mode=padding_mode,
+            dilation=dilation,
+            groups=groups,
+            weight_attr=weight_attr,
+            bias_attr=bias_attr,
+            data_format=data_format)
+        self.valid_trigx = valid_trigx
+        self.valid_trigy = valid_trigy
+
+    def _conv_forward(self,
+                      input: paddle.Tensor,
+                      weight: paddle.Tensor,
+                      bias: Optional[paddle.Tensor]):
+        validx, validy = 0, 0
+        if self.valid_trigx:
+            validx = (input.shape[-2] *
+                      (self._stride[-2] - 1) - 1 + self._kernel_size[-2]) // 2
+        if self.valid_trigy:
+            validy = (input.shape[-1] *
+                      (self._stride[-1] - 1) - 1 + self._kernel_size[-1]) // 2
+        return F.conv2d(input, weight, bias, self._stride, (validx, validy),
+                        self._dilation, self._groups)
+
+    def forward(self, input: paddle.Tensor) -> paddle.Tensor:
+        return self._conv_forward(input, self.weight, self.bias)

paddlespeech/s2t/modules/encoder.py

@@ -14,7 +14,10 @@
 # limitations under the License.
 # Modified from wenet(https://github.com/wenet-e2e/wenet)
 """Encoder definition."""
+from typing import List
+from typing import Optional
 from typing import Tuple
+from typing import Union
 
 import paddle
 from paddle import nn
@@ -22,6 +25,7 @@ from typeguard import check_argument_types
 
 from paddlespeech.s2t.modules.activation import get_activation
 from paddlespeech.s2t.modules.align import LayerNorm
+from paddlespeech.s2t.modules.align import Linear
 from paddlespeech.s2t.modules.attention import MultiHeadedAttention
 from paddlespeech.s2t.modules.attention import RelPositionMultiHeadedAttention
 from paddlespeech.s2t.modules.conformer_convolution import ConvolutionModule
@@ -29,6 +33,7 @@ from paddlespeech.s2t.modules.embedding import NoPositionalEncoding
 from paddlespeech.s2t.modules.embedding import PositionalEncoding
 from paddlespeech.s2t.modules.embedding import RelPositionalEncoding
 from paddlespeech.s2t.modules.encoder_layer import ConformerEncoderLayer
+from paddlespeech.s2t.modules.encoder_layer import SqueezeformerEncoderLayer
 from paddlespeech.s2t.modules.encoder_layer import TransformerEncoderLayer
 from paddlespeech.s2t.modules.mask import add_optional_chunk_mask
 from paddlespeech.s2t.modules.mask import make_non_pad_mask
@@ -36,12 +41,19 @@ from paddlespeech.s2t.modules.positionwise_feed_forward import PositionwiseFeedF
 from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling4
 from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling6
 from paddlespeech.s2t.modules.subsampling import Conv2dSubsampling8
+from paddlespeech.s2t.modules.subsampling import DepthwiseConv2DSubsampling4
 from paddlespeech.s2t.modules.subsampling import LinearNoSubsampling
+from paddlespeech.s2t.modules.time_reduction import TimeReductionLayer1D
+from paddlespeech.s2t.modules.time_reduction import TimeReductionLayer2D
+from paddlespeech.s2t.modules.time_reduction import TimeReductionLayerStream
 from paddlespeech.s2t.utils.log import Log
 
 logger = Log(__name__).getlog()
 
-__all__ = ["BaseEncoder", 'TransformerEncoder', "ConformerEncoder"]
+__all__ = [
+    "BaseEncoder", 'TransformerEncoder', "ConformerEncoder",
+    "SqueezeformerEncoder"
+]
 
 
 class BaseEncoder(nn.Layer):
@@ -487,3 +499,366 @@ class ConformerEncoder(BaseEncoder):
                 normalize_before=normalize_before,
                 concat_after=concat_after) for _ in range(num_blocks)
         ])
+
+
+class SqueezeformerEncoder(nn.Layer):
+    def __init__(self,
+                 input_size: int,
+                 encoder_dim: int=256,
+                 output_size: int=256,
+                 attention_heads: int=4,
+                 num_blocks: int=12,
+                 reduce_idx: Optional[Union[int, List[int]]]=5,
+                 recover_idx: Optional[Union[int, List[int]]]=11,
+                 feed_forward_expansion_factor: int=4,
+                 dw_stride: bool=False,
+                 input_dropout_rate: float=0.1,
+                 pos_enc_layer_type: str="rel_pos",
+                 time_reduction_layer_type: str="conv1d",
+                 feed_forward_dropout_rate: float=0.1,
+                 attention_dropout_rate: float=0.1,
+                 cnn_module_kernel: int=31,
+                 cnn_norm_type: str="layer_norm",
+                 dropout: float=0.1,
+                 causal: bool=False,
+                 adaptive_scale: bool=True,
+                 activation_type: str="swish",
+                 init_weights: bool=True,
+                 global_cmvn: paddle.nn.Layer=None,
+                 normalize_before: bool=False,
+                 use_dynamic_chunk: bool=False,
+                 concat_after: bool=False,
+                 static_chunk_size: int=0,
+                 use_dynamic_left_chunk: bool=False):
+        """Construct SqueezeformerEncoder
+
+        Args:
+            input_size to use_dynamic_chunk, see in Transformer BaseEncoder.
+            encoder_dim (int): The hidden dimension of encoder layer.
+            output_size (int): The output dimension of final projection layer.
+            attention_heads (int): Num of attention head in attention module.
+            num_blocks (int): Num of encoder layers.
+            reduce_idx Optional[Union[int, List[int]]]:
+                reduce layer index, from 40ms to 80ms per frame.
+            recover_idx Optional[Union[int, List[int]]]:
+                recover layer index, from 80ms to 40ms per frame.
+            feed_forward_expansion_factor (int): Enlarge coefficient of FFN.
+            dw_stride (bool): Whether do depthwise convolution
+                              on subsampling module.
+            input_dropout_rate (float): Dropout rate of input projection layer.
+            pos_enc_layer_type (str): Self attention type.
+            time_reduction_layer_type (str): Conv1d or Conv2d reduction layer.
+            cnn_module_kernel (int): Kernel size of CNN module.
+            activation_type (str): Encoder activation function type.
+            cnn_module_kernel (int): Kernel size of convolution module.
+            adaptive_scale (bool): Whether to use adaptive scale.
+            init_weights (bool): Whether to initialize weights.
+            causal (bool): whether to use causal convolution or not.
+        """
+        assert check_argument_types()
+        super().__init__()
+        self.global_cmvn = global_cmvn
+        self.reduce_idx: Optional[Union[int, List[int]]] = [reduce_idx] \
+            if type(reduce_idx) == int else reduce_idx
+        self.recover_idx: Optional[Union[int, List[int]]] = [recover_idx] \
+            if type(recover_idx) == int else recover_idx
+        self.check_ascending_list()
+        if reduce_idx is None:
+            self.time_reduce = None
+        else:
+            if recover_idx is None:
+                self.time_reduce = 'normal'  # no recovery at the end
+            else:
+                self.time_reduce = 'recover'  # recovery at the end
+                assert len(self.reduce_idx) == len(self.recover_idx)
+            self.reduce_stride = 2
+        self._output_size = output_size
+        self.normalize_before = normalize_before
+        self.static_chunk_size = static_chunk_size
+        self.use_dynamic_chunk = use_dynamic_chunk
+        self.use_dynamic_left_chunk = use_dynamic_left_chunk
+        activation = get_activation(activation_type)
+
+        # self-attention module definition
+        if pos_enc_layer_type != "rel_pos":
+            encoder_selfattn_layer = MultiHeadedAttention
+            encoder_selfattn_layer_args = (attention_heads, output_size,
+                                           attention_dropout_rate)
+        else:
+            encoder_selfattn_layer = RelPositionMultiHeadedAttention
+            encoder_selfattn_layer_args = (attention_heads, encoder_dim,
+                                           attention_dropout_rate,
+                                           adaptive_scale, init_weights)
+
+        # feed-forward module definition
+        positionwise_layer = PositionwiseFeedForward
+        positionwise_layer_args = (
+            encoder_dim, encoder_dim * feed_forward_expansion_factor,
+            feed_forward_dropout_rate, activation, adaptive_scale, init_weights)
+
+        # convolution module definition
+        convolution_layer = ConvolutionModule
+        convolution_layer_args = (encoder_dim, cnn_module_kernel, activation,
+                                  cnn_norm_type, causal, True, adaptive_scale,
+                                  init_weights)
+
+        self.embed = DepthwiseConv2DSubsampling4(
+            1, encoder_dim,
+            RelPositionalEncoding(encoder_dim, dropout_rate=0.1), dw_stride,
+            input_size, input_dropout_rate, init_weights)
+
+        self.preln = LayerNorm(encoder_dim)
+        self.encoders = paddle.nn.LayerList([
+            SqueezeformerEncoderLayer(
+                encoder_dim,
+                encoder_selfattn_layer(*encoder_selfattn_layer_args),
+                positionwise_layer(*positionwise_layer_args),
+                convolution_layer(*convolution_layer_args),
+                positionwise_layer(*positionwise_layer_args), normalize_before,
+                dropout, concat_after) for _ in range(num_blocks)
+        ])
+        if time_reduction_layer_type == 'conv1d':
+            time_reduction_layer = TimeReductionLayer1D
+            time_reduction_layer_args = {
+                'channel': encoder_dim,
+                'out_dim': encoder_dim,
+            }
+        elif time_reduction_layer_type == 'stream':
+            time_reduction_layer = TimeReductionLayerStream
+            time_reduction_layer_args = {
+                'channel': encoder_dim,
+                'out_dim': encoder_dim,
+            }
+        else:
+            time_reduction_layer = TimeReductionLayer2D
+            time_reduction_layer_args = {'encoder_dim': encoder_dim}
+
+        self.time_reduction_layer = time_reduction_layer(
+            **time_reduction_layer_args)
+        self.time_recover_layer = Linear(encoder_dim, encoder_dim)
+        self.final_proj = None
+        if output_size != encoder_dim:
+            self.final_proj = Linear(encoder_dim, output_size)
+
+    def output_size(self) -> int:
+        return self._output_size
+
+    def forward(
+            self,
+            xs: paddle.Tensor,
+            xs_lens: paddle.Tensor,
+            decoding_chunk_size: int=0,
+            num_decoding_left_chunks: int=-1,
+    ) -> Tuple[paddle.Tensor, paddle.Tensor]:
+        """Embed positions in tensor.
+        Args:
+            xs: padded input tensor (B, L, 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, lens and mask
+        """
+        masks = make_non_pad_mask(xs_lens).unsqueeze(1)  # (B, 1, L)
+
+        if self.global_cmvn is not None:
+            xs = self.global_cmvn(xs)
+        xs, pos_emb, masks = self.embed(xs, masks)
+        mask_pad = masks
+        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)
+        xs_lens = chunk_masks.squeeze(1).sum(1)
+        xs = self.preln(xs)
+        recover_activations: \
+            List[Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor, paddle.Tensor]] = []
+        index = 0
+        for i, layer in enumerate(self.encoders):
+            if self.reduce_idx is not None:
+                if self.time_reduce is not None and i in self.reduce_idx:
+                    recover_activations.append(
+                        (xs, chunk_masks, pos_emb, mask_pad))
+                    xs, xs_lens, chunk_masks, mask_pad = self.time_reduction_layer(
+                        xs, xs_lens, chunk_masks, mask_pad)
+                    pos_emb = pos_emb[:, ::2, :]
+                    index += 1
+
+            if self.recover_idx is not None:
+                if self.time_reduce == 'recover' and i in self.recover_idx:
+                    index -= 1
+                    recover_tensor, recover_chunk_masks, recover_pos_emb, recover_mask_pad = recover_activations[
+                        index]
+                    # recover output length for ctc decode
+                    xs = paddle.repeat_interleave(xs, repeats=2, axis=1)
+                    xs = self.time_recover_layer(xs)
+                    recoverd_t = recover_tensor.shape[1]
+                    xs = recover_tensor + xs[:, :recoverd_t, :]
+                    chunk_masks = recover_chunk_masks
+                    pos_emb = recover_pos_emb
+                    mask_pad = recover_mask_pad
+
+            xs, chunk_masks, _, _ = layer(xs, chunk_masks, pos_emb, mask_pad)
+
+        if self.final_proj is not None:
+            xs = self.final_proj(xs)
+        return xs, masks
+
+    def check_ascending_list(self):
+        if self.reduce_idx is not None:
+            assert self.reduce_idx == sorted(self.reduce_idx), \
+                "reduce_idx should be int or ascending list"
+        if self.recover_idx is not None:
+            assert self.recover_idx == sorted(self.recover_idx), \
+                "recover_idx should be int or ascending list"
+
+    def calculate_downsampling_factor(self, i: int) -> int:
+        if self.reduce_idx is None:
+            return 1
+        else:
+            reduce_exp, recover_exp = 0, 0
+            for exp, rd_idx in enumerate(self.reduce_idx):
+                if i >= rd_idx:
+                    reduce_exp = exp + 1
+            if self.recover_idx is not None:
+                for exp, rc_idx in enumerate(self.recover_idx):
+                    if i >= rc_idx:
+                        recover_exp = exp + 1
+            return int(2**(reduce_exp - recover_exp))
+
+    def forward_chunk(
+            self,
+            xs: paddle.Tensor,
+            offset: int,
+            required_cache_size: int,
+            att_cache: paddle.Tensor=paddle.zeros([0, 0, 0, 0]),
+            cnn_cache: paddle.Tensor=paddle.zeros([0, 0, 0, 0]),
+            att_mask: paddle.Tensor=paddle.ones([0, 0, 0], dtype=paddle.bool),
+    ) -> Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor]:
+        """ Forward just one chunk
+
+        Args:
+            xs (paddle.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 (paddle.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 (paddle.Tensor): cache tensor for cnn_module in conformer,
+                (elayers, b=1, hidden-dim, cache_t2), where
+                `cache_t2 == cnn.lorder - 1`
+
+        Returns:
+            paddle.Tensor: output of current input xs,
+                with shape (b=1, chunk_size, hidden-dim).
+            paddle.Tensor: new attention cache required for next chunk, with
+                dynamic shape (elayers, head, ?, d_k * 2)
+                depending on required_cache_size.
+            paddle.Tensor: new conformer cnn cache required for next chunk, with
+                same shape as the original cnn_cache.
+        """
+        assert xs.shape[0] == 1  # batch size must be one
+
+        if self.global_cmvn is not None:
+            xs = self.global_cmvn(xs)
+
+        # tmp_masks is just for interface compatibility, [B=1, C=1, T]
+        tmp_masks = paddle.ones([1, 1, xs.shape[1]], dtype=paddle.bool)
+        # before embed, xs=(B, T, D1), pos_emb=(B=1, T, D)
+        xs, pos_emb, _ = self.embed(xs, tmp_masks, offset=offset)
+
+        # NOTE(xcsong): After  embed, shape(xs) is (b=1, chunk_size, hidden-dim)
+        elayers, cache_t1 = att_cache.shape[0], att_cache.shape[2]
+        chunk_size = xs.shape[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 = []
+
+        mask_pad = paddle.ones([1, xs.shape[1]], dtype=paddle.bool)
+        mask_pad = mask_pad.unsqueeze(1)
+        max_att_len: int = 0
+        recover_activations: \
+            List[Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor, paddle.Tensor]] = []
+        index = 0
+        xs_lens = paddle.to_tensor([xs.shape[1]], dtype=paddle.int32)
+        xs = self.preln(xs)
+        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)
+            if self.reduce_idx is not None:
+                if self.time_reduce is not None and i in self.reduce_idx:
+                    recover_activations.append(
+                        (xs, att_mask, pos_emb, mask_pad))
+                    xs, xs_lens, att_mask, mask_pad = self.time_reduction_layer(
+                        xs, xs_lens, att_mask, mask_pad)
+                    pos_emb = pos_emb[:, ::2, :]
+                    index += 1
+
+            if self.recover_idx is not None:
+                if self.time_reduce == 'recover' and i in self.recover_idx:
+                    index -= 1
+                    recover_tensor, recover_att_mask, recover_pos_emb, recover_mask_pad = recover_activations[
+                        index]
+                    # recover output length for ctc decode
+                    xs = paddle.repeat_interleave(xs, repeats=2, axis=1)
+                    xs = self.time_recover_layer(xs)
+                    recoverd_t = recover_tensor.shape[1]
+                    xs = recover_tensor + xs[:, :recoverd_t, :]
+                    att_mask = recover_att_mask
+                    pos_emb = recover_pos_emb
+                    mask_pad = recover_mask_pad
+
+            factor = self.calculate_downsampling_factor(i)
+            att_cache1 = att_cache[
+                i:i + 1][:, :, ::factor, :][:, :, :pos_emb.shape[1] - xs.shape[
+                    1], :]
+            cnn_cache1 = cnn_cache[i] if cnn_cache.shape[0] > 0 else cnn_cache
+            xs, _, new_att_cache, new_cnn_cache = layer(
+                xs,
+                att_mask,
+                pos_emb,
+                att_cache=att_cache1,
+                cnn_cache=cnn_cache1)
+            # 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)
+            cached_att = new_att_cache[:, :, next_cache_start // factor:, :]
+            cached_cnn = new_cnn_cache.unsqueeze(0)
+            cached_att = cached_att.repeat_interleave(repeats=factor, axis=2)
+            if i == 0:
+                # record length for the first block as max length
+                max_att_len = cached_att.shape[2]
+            r_att_cache.append(cached_att[:, :, :max_att_len, :])
+            r_cnn_cache.append(cached_cnn)
+        # 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 = paddle.concat(r_att_cache, axis=0)
+        # NOTE(xcsong): shape(r_cnn_cache) is (e, b=1, hidden-dim, cache_t2)
+        r_cnn_cache = paddle.concat(r_cnn_cache, axis=0)
+
+        if self.final_proj is not None:
+            xs = self.final_proj(xs)
+        return xs, r_att_cache, r_cnn_cache

paddlespeech/s2t/modules/encoder_layer.py

@@ -26,7 +26,10 @@ from paddlespeech.s2t.utils.log import Log
 
 logger = Log(__name__).getlog()
 
-__all__ = ["TransformerEncoderLayer", "ConformerEncoderLayer"]
+__all__ = [
+    "TransformerEncoderLayer", "ConformerEncoderLayer",
+    "SqueezeformerEncoderLayer"
+]
 
 
 class TransformerEncoderLayer(nn.Layer):
@@ -276,3 +279,125 @@ class ConformerEncoderLayer(nn.Layer):
             x = self.norm_final(x)
 
         return x, mask, new_att_cache, new_cnn_cache
+
+
+class SqueezeformerEncoderLayer(nn.Layer):
+    """Encoder layer module."""
+
+    def __init__(self,
+                 size: int,
+                 self_attn: paddle.nn.Layer,
+                 feed_forward1: Optional[nn.Layer]=None,
+                 conv_module: Optional[nn.Layer]=None,
+                 feed_forward2: Optional[nn.Layer]=None,
+                 normalize_before: bool=False,
+                 dropout_rate: float=0.1,
+                 concat_after: bool=False):
+        """Construct an EncoderLayer object.
+
+        Args:
+            size (int): Input dimension.
+            self_attn (paddle.nn.Layer): Self-attention module instance.
+                `MultiHeadedAttention` or `RelPositionMultiHeadedAttention`
+                instance can be used as the argument.
+            feed_forward1 (paddle.nn.Layer): Feed-forward module instance.
+                `PositionwiseFeedForward` instance can be used as the argument.
+            conv_module (paddle.nn.Layer): Convolution module instance.
+                `ConvlutionLayer` instance can be used as the argument.
+            feed_forward2 (paddle.nn.Layer): Feed-forward module instance.
+                `PositionwiseFeedForward` instance can be used as the argument.
+            dropout_rate (float): Dropout rate.
+            normalize_before (bool):
+                True: use layer_norm before each sub-block.
+                False: use layer_norm after each sub-block.
+        """
+        super().__init__()
+        self.size = size
+        self.self_attn = self_attn
+        self.layer_norm1 = LayerNorm(size)
+        self.ffn1 = feed_forward1
+        self.layer_norm2 = LayerNorm(size)
+        self.conv_module = conv_module
+        self.layer_norm3 = LayerNorm(size)
+        self.ffn2 = feed_forward2
+        self.layer_norm4 = LayerNorm(size)
+        self.normalize_before = normalize_before
+        self.dropout = nn.Dropout(dropout_rate)
+        self.concat_after = concat_after
+        if concat_after:
+            self.concat_linear = Linear(size + size, size)
+        else:
+            self.concat_linear = nn.Identity()
+
+    def forward(
+            self,
+            x: paddle.Tensor,
+            mask: paddle.Tensor,
+            pos_emb: paddle.Tensor,
+            mask_pad: paddle.Tensor=paddle.ones([0, 0, 0], dtype=paddle.bool),
+            att_cache: paddle.Tensor=paddle.zeros([0, 0, 0, 0]),
+            cnn_cache: paddle.Tensor=paddle.zeros([0, 0, 0, 0]),
+    ) -> Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor, paddle.Tensor]:
+        """Compute encoded features.
+        Args:
+            x (paddle.Tensor): Input tensor (#batch, time, size).
+            mask (paddle.Tensor): Mask tensor for the input (#batch, time, time).
+                (0,0,0) means fake mask.
+            pos_emb (paddle.Tensor): postional encoding, must not be None
+                for ConformerEncoderLayer
+            mask_pad (paddle.Tensor): batch padding mask used for conv module.
+               (#batch, 1，time), (0, 0, 0) means fake mask.
+            att_cache (paddle.Tensor): Cache tensor of the KEY & VALUE
+                (#batch=1, head, cache_t1, d_k * 2), head * d_k == size.
+            cnn_cache (paddle.Tensor): Convolution cache in conformer layer
+                (1, #batch=1, size, cache_t2). First dim will not be used, just
+                for dy2st.
+        Returns:
+           paddle.Tensor: Output tensor (#batch, time, size).
+           paddle.Tensor: Mask tensor (#batch, time, time).
+           paddle.Tensor: att_cache tensor,
+                (#batch=1, head, cache_t1 + time, d_k * 2).
+           paddle.Tensor: cnn_cahce tensor (#batch, size, cache_t2).
+        """
+        # self attention module
+        residual = x
+        if self.normalize_before:
+            x = self.layer_norm1(x)
+        x_att, new_att_cache = self.self_attn(x, x, x, mask, pos_emb, att_cache)
+        if self.concat_after:
+            x_concat = paddle.concat((x, x_att), axis=-1)
+            x = residual + self.concat_linear(x_concat)
+        else:
+            x = residual + self.dropout(x_att)
+        if not self.normalize_before:
+            x = self.layer_norm1(x)
+
+        # ffn module
+        residual = x
+        if self.normalize_before:
+            x = self.layer_norm2(x)
+        x = self.ffn1(x)
+        x = residual + self.dropout(x)
+        if not self.normalize_before:
+            x = self.layer_norm2(x)
+
+        # conv module
+        residual = x
+        if self.normalize_before:
+            x = self.layer_norm3(x)
+        x, new_cnn_cache = self.conv_module(x, mask_pad, cnn_cache)
+        x = residual + self.dropout(x)
+        if not self.normalize_before:
+            x = self.layer_norm3(x)
+
+        # ffn module
+        residual = x
+        if self.normalize_before:
+            x = self.layer_norm4(x)
+        x = self.ffn2(x)
+        # we do not use dropout here since it is inside feed forward function
+        x = residual + self.dropout(x)
+        if not self.normalize_before:
+            x = self.layer_norm4(x)
+
+        return x, mask, new_att_cache, new_cnn_cache

paddlespeech/s2t/modules/positionwise_feed_forward.py

@@ -16,6 +16,7 @@
 """Positionwise feed forward layer definition."""
 import paddle
 from paddle import nn
+from paddle.nn import initializer as I
 
 from paddlespeech.s2t.modules.align import Linear
 from paddlespeech.s2t.utils.log import Log
@@ -32,7 +33,9 @@ class PositionwiseFeedForward(nn.Layer):
                  idim: int,
                  hidden_units: int,
                  dropout_rate: float,
-                 activation: nn.Layer=nn.ReLU()):
+                 activation: nn.Layer=nn.ReLU(),
+                 adaptive_scale: bool=False,
+                 init_weights: bool=False):
         """Construct a PositionwiseFeedForward object.
 
         FeedForward are appied on each position of the sequence.
@@ -45,10 +48,35 @@ class PositionwiseFeedForward(nn.Layer):
             activation (paddle.nn.Layer): Activation function
         """
         super().__init__()
+        self.idim = idim
+        self.hidden_units = hidden_units
         self.w_1 = Linear(idim, hidden_units)
         self.activation = activation
         self.dropout = nn.Dropout(dropout_rate)
         self.w_2 = Linear(hidden_units, idim)
+        self.adaptive_scale = adaptive_scale
+        if self.adaptive_scale:
+            ada_scale = self.create_parameter(
+                [1, 1, idim], default_initializer=I.XavierUniform())
+            self.add_parameter('ada_scale', ada_scale)
+            ada_bias = self.create_parameter(
+                [1, 1, idim], default_initializer=I.XavierUniform())
+            self.add_parameter('ada_bias', ada_bias)
+
+        if init_weights:
+            self.init_weights()
+
+    def init_weights(self):
+        ffn1_max = self.idim**-0.5
+        ffn2_max = self.hidden_units**-0.5
+        self.w_1._param_attr = paddle.nn.initializer.Uniform(
+            low=-ffn1_max, high=ffn1_max)
+        self.w_1._bias_attr = paddle.nn.initializer.Uniform(
+            low=-ffn1_max, high=ffn1_max)
+        self.w_2._param_attr = paddle.nn.initializer.Uniform(
+            low=-ffn2_max, high=ffn2_max)
+        self.w_2._bias_attr = paddle.nn.initializer.Uniform(
+            low=-ffn2_max, high=ffn2_max)
 
     def forward(self, xs: paddle.Tensor) -> paddle.Tensor:
         """Forward function.
@@ -57,4 +85,6 @@ class PositionwiseFeedForward(nn.Layer):
         Returns:
             output tensor, (B, Lmax, D)
         """
+        if self.adaptive_scale:
+            xs = self.ada_scale * xs + self.ada_bias
         return self.w_2(self.dropout(self.activation(self.w_1(xs))))

paddlespeech/s2t/modules/subsampling.py

@@ -29,7 +29,7 @@ logger = Log(__name__).getlog()
 
 __all__ = [
     "LinearNoSubsampling", "Conv2dSubsampling4", "Conv2dSubsampling6",
-    "Conv2dSubsampling8"
+    "Conv2dSubsampling8", "DepthwiseConv2DSubsampling4"
 ]
 
 
@@ -249,3 +249,67 @@ class Conv2dSubsampling8(Conv2dSubsampling):
         x = self.linear(x.transpose([0, 2, 1, 3]).reshape([b, -1, c * f]))
         x, pos_emb = self.pos_enc(x, offset)
         return x, pos_emb, x_mask[:, :, :-2:2][:, :, :-2:2][:, :, :-2:2]
+
+
+class DepthwiseConv2DSubsampling4(BaseSubsampling):
+    """Depthwise Convolutional 2D subsampling (to 1/4 length).
+
+        Args:
+            idim (int): Input dimension.
+            odim (int): Output dimension.
+            pos_enc_class (nn.Layer): position encoding class.
+            dw_stride (int): Whether do depthwise convolution.
+            input_size (int): filter bank dimension.
+
+        """
+
+    def __init__(self,
+                 idim: int,
+                 odim: int,
+                 pos_enc_class: nn.Layer,
+                 dw_stride: bool=False,
+                 input_size: int=80,
+                 input_dropout_rate: float=0.1,
+                 init_weights: bool=True):
+        super(DepthwiseConv2DSubsampling4, self).__init__()
+        self.idim = idim
+        self.odim = odim
+        self.pw_conv = Conv2D(
+            in_channels=idim, out_channels=odim, kernel_size=3, stride=2)
+        self.act1 = nn.ReLU()
+        self.dw_conv = Conv2D(
+            in_channels=odim,
+            out_channels=odim,
+            kernel_size=3,
+            stride=2,
+            groups=odim if dw_stride else 1)
+        self.act2 = nn.ReLU()
+        self.pos_enc = pos_enc_class
+        self.input_proj = nn.Sequential(
+            Linear(odim * (((input_size - 1) // 2 - 1) // 2), odim),
+            nn.Dropout(p=input_dropout_rate))
+        if init_weights:
+            linear_max = (odim * input_size / 4)**-0.5
+            self.input_proj.state_dict()[
+                '0.weight'] = paddle.nn.initializer.Uniform(
+                    low=-linear_max, high=linear_max)
+            self.input_proj.state_dict()[
+                '0.bias'] = paddle.nn.initializer.Uniform(
+                    low=-linear_max, high=linear_max)
+
+        self.subsampling_rate = 4
+        # 6 = (3 - 1) * 1 + (3 - 1) * 2
+        self.right_context = 6
+
+    def forward(self, x: paddle.Tensor, x_mask: paddle.Tensor, offset: int=0
+                ) -> Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor]:
+        x = x.unsqueeze(1)  # (b, c=1, t, f)
+        x = self.pw_conv(x)
+        x = self.act1(x)
+        x = self.dw_conv(x)
+        x = self.act2(x)
+        b, c, t, f = x.shape
+        x = x.transpose([0, 2, 1, 3]).reshape([b, -1, c * f])
+        x, pos_emb = self.pos_enc(x, offset)
+        x = self.input_proj(x)
+        return x, pos_emb, x_mask[:, :, :-2:2][:, :, :-2:2]

paddlespeech/s2t/modules/time_reduction.py

@@ -0,0 +1,263 @@
+# Copyright (c) 2021 PaddlePaddle Authors. All Rights Reserved.
+# Copyright 2019 Mobvoi Inc. All Rights Reserved.
+#
+# 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 wenet(https://github.com/wenet-e2e/wenet)
+"""Subsampling layer definition."""
+from typing import Tuple
+
+import paddle
+import paddle.nn.functional as F
+from paddle import nn
+
+from paddlespeech.s2t import masked_fill
+from paddlespeech.s2t.modules.align import Conv1D
+from paddlespeech.s2t.modules.conv2d import Conv2DValid
+from paddlespeech.s2t.utils.log import Log
+
+logger = Log(__name__).getlog()
+
+__all__ = [
+    "TimeReductionLayerStream", "TimeReductionLayer1D", "TimeReductionLayer2D"
+]
+
+
+class TimeReductionLayer1D(nn.Layer):
+    """
+    Modified NeMo,
+    Squeezeformer Time Reduction procedure.
+    Downsamples the audio by `stride` in the time dimension.
+    Args:
+        channel (int): input dimension of
+                       MultiheadAttentionMechanism and PositionwiseFeedForward
+        out_dim (int): Output dimension of the module.
+        kernel_size (int): Conv kernel size for
+                           depthwise convolution in convolution module
+        stride (int): Downsampling factor in time dimension.
+    """
+
+    def __init__(self,
+                 channel: int,
+                 out_dim: int,
+                 kernel_size: int=5,
+                 stride: int=2):
+        super(TimeReductionLayer1D, self).__init__()
+
+        self.channel = channel
+        self.out_dim = out_dim
+        self.kernel_size = kernel_size
+        self.stride = stride
+        self.padding = max(0, self.kernel_size - self.stride)
+
+        self.dw_conv = Conv1D(
+            in_channels=channel,
+            out_channels=channel,
+            kernel_size=kernel_size,
+            stride=stride,
+            padding=self.padding,
+            groups=channel, )
+
+        self.pw_conv = Conv1D(
+            in_channels=channel,
+            out_channels=out_dim,
+            kernel_size=1,
+            stride=1,
+            padding=0,
+            groups=1, )
+
+        self.init_weights()
+
+    def init_weights(self):
+        dw_max = self.kernel_size**-0.5
+        pw_max = self.channel**-0.5
+        self.dw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.dw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.pw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+        self.pw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+
+    def forward(
+            self,
+            xs,
+            xs_lens: paddle.Tensor,
+            mask: paddle.Tensor=paddle.ones((0, 0, 0), dtype=paddle.bool),
+            mask_pad: paddle.Tensor=paddle.ones((0, 0, 0),
+                                                dtype=paddle.bool), ):
+        xs = xs.transpose([0, 2, 1])  # [B, C, T]
+        xs = masked_fill(xs, mask_pad.equal(0), 0.0)
+
+        xs = self.dw_conv(xs)
+        xs = self.pw_conv(xs)
+
+        xs = xs.transpose([0, 2, 1])  # [B, T, C]
+
+        B, T, D = xs.shape
+        mask = mask[:, ::self.stride, ::self.stride]
+        mask_pad = mask_pad[:, :, ::self.stride]
+        L = mask_pad.shape[-1]
+        # For JIT exporting, we remove F.pad operator.
+        if L - T < 0:
+            xs = xs[:, :L - T, :]
+        else:
+            dummy_pad = paddle.zeros([B, L - T, D], dtype=paddle.float32)
+            xs = paddle.concat([xs, dummy_pad], axis=1)
+
+        xs_lens = (xs_lens + 1) // 2
+        return xs, xs_lens, mask, mask_pad
+
+
+class TimeReductionLayer2D(nn.Layer):
+    def __init__(self, kernel_size: int=5, stride: int=2, encoder_dim: int=256):
+        super(TimeReductionLayer2D, self).__init__()
+        self.encoder_dim = encoder_dim
+        self.kernel_size = kernel_size
+        self.dw_conv = Conv2DValid(
+            in_channels=encoder_dim,
+            out_channels=encoder_dim,
+            kernel_size=(kernel_size, 1),
+            stride=stride,
+            valid_trigy=True)
+        self.pw_conv = Conv2DValid(
+            in_channels=encoder_dim,
+            out_channels=encoder_dim,
+            kernel_size=1,
+            stride=1,
+            valid_trigx=False,
+            valid_trigy=False)
+
+        self.kernel_size = kernel_size
+        self.stride = stride
+        self.init_weights()
+
+    def init_weights(self):
+        dw_max = self.kernel_size**-0.5
+        pw_max = self.encoder_dim**-0.5
+        self.dw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.dw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.pw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+        self.pw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+
+    def forward(
+            self,
+            xs: paddle.Tensor,
+            xs_lens: paddle.Tensor,
+            mask: paddle.Tensor=paddle.ones((0, 0, 0), dtype=paddle.bool),
+            mask_pad: paddle.Tensor=paddle.ones((0, 0, 0), dtype=paddle.bool),
+    ) -> Tuple[paddle.Tensor, paddle.Tensor, paddle.Tensor, paddle.Tensor]:
+        xs = masked_fill(xs, mask_pad.transpose([0, 2, 1]).equal(0), 0.0)
+        xs = xs.unsqueeze(1)
+        padding1 = self.kernel_size - self.stride
+        xs = F.pad(
+            xs, (0, 0, 0, 0, 0, padding1, 0, 0), mode='constant', value=0.)
+        xs = self.dw_conv(xs.transpose([0, 3, 2, 1]))
+        xs = self.pw_conv(xs).transpose([0, 3, 2, 1]).squeeze(1)
+        tmp_length = xs.shape[1]
+        xs_lens = (xs_lens + 1) // 2
+        padding2 = max(0, (xs_lens.max() - tmp_length).item())
+        batch_size, hidden = xs.shape[0], xs.shape[-1]
+        dummy_pad = paddle.zeros(
+            [batch_size, padding2, hidden], dtype=paddle.float32)
+        xs = paddle.concat([xs, dummy_pad], axis=1)
+        mask = mask[:, ::2, ::2]
+        mask_pad = mask_pad[:, :, ::2]
+        return xs, xs_lens, mask, mask_pad
+
+
+class TimeReductionLayerStream(nn.Layer):
+    """
+    Squeezeformer Time Reduction procedure.
+    Downsamples the audio by `stride` in the time dimension.
+    Args:
+        channel (int): input dimension of
+            MultiheadAttentionMechanism and PositionwiseFeedForward
+        out_dim (int): Output dimension of the module.
+        kernel_size (int): Conv kernel size for
+            depthwise convolution in convolution module
+        stride (int): Downsampling factor in time dimension.
+    """
+
+    def __init__(self,
+                 channel: int,
+                 out_dim: int,
+                 kernel_size: int=1,
+                 stride: int=2):
+        super(TimeReductionLayerStream, self).__init__()
+
+        self.channel = channel
+        self.out_dim = out_dim
+        self.kernel_size = kernel_size
+        self.stride = stride
+
+        self.dw_conv = Conv1D(
+            in_channels=channel,
+            out_channels=channel,
+            kernel_size=kernel_size,
+            stride=stride,
+            padding=0,
+            groups=channel)
+
+        self.pw_conv = Conv1D(
+            in_channels=channel,
+            out_channels=out_dim,
+            kernel_size=1,
+            stride=1,
+            padding=0,
+            groups=1)
+        self.init_weights()
+
+    def init_weights(self):
+        dw_max = self.kernel_size**-0.5
+        pw_max = self.channel**-0.5
+        self.dw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.dw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-dw_max, high=dw_max)
+        self.pw_conv._param_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+        self.pw_conv._bias_attr = paddle.nn.initializer.Uniform(
+            low=-pw_max, high=pw_max)
+
+    def forward(
+            self,
+            xs,
+            xs_lens: paddle.Tensor,
+            mask: paddle.Tensor=paddle.ones([0, 0, 0], dtype=paddle.bool),
+            mask_pad: paddle.Tensor=paddle.ones([0, 0, 0], dtype=paddle.bool)):
+        xs = xs.transpose([0, 2, 1])  # [B, C, T]
+        xs = masked_fill(xs, mask_pad.equal(0), 0.0)
+
+        xs = self.dw_conv(xs)
+        xs = self.pw_conv(xs)
+
+        xs = xs.transpose([0, 2, 1])  # [B, T, C]
+
+        B, T, D = xs.shape
+        mask = mask[:, ::self.stride, ::self.stride]
+        mask_pad = mask_pad[:, :, ::self.stride]
+        L = mask_pad.shape[-1]
+        # For JIT exporting, we remove F.pad operator.
+        if L - T < 0:
+            xs = xs[:, :L - T, :]
+        else:
+            dummy_pad = paddle.zeros([B, L - T, D], dtype=paddle.float32)
+            xs = paddle.concat([xs, dummy_pad], axis=1)
+
+        xs_lens = (xs_lens + 1) // 2
+        return xs, xs_lens, mask, mask_pad

paddlespeech/s2t/utils/utility.py

@@ -130,8 +130,11 @@ def get_subsample(config):
     Returns:
         int: subsample rate.
     """
-    input_layer = config["encoder_conf"]["input_layer"]
+    if config['encoder'] == 'squeezeformer':
-    assert input_layer in ["conv2d", "conv2d6", "conv2d8"]
+        return 4
+    else:
+        input_layer = config["encoder_conf"]["input_layer"]
+        assert input_layer in ["conv2d", "conv2d6", "conv2d8"]
     if input_layer == "conv2d":
         return 4
     elif input_layer == "conv2d6":