fix the bug of the relative position encoding.

egs/libri_trans/asr/conf/base.yaml

@@ -5,8 +5,9 @@ clip-norm: 10.0
 lr-scheduler: inverse_sqrt
 lr-scheduler: inverse_sqrt
 warmup-init-lr: 1e-7
 warmup-init-lr: 1e-7
 warmup-updates: 10000
 warmup-updates: 10000
+weight-decay: 1e-6
 lr: 2e-3
 lr: 2e-3
-#adam_betas: (0.9,0.98)
+adam_betas: (0.9,0.98)
 
 
 criterion: label_smoothed_cross_entropy_with_ctc
 criterion: label_smoothed_cross_entropy_with_ctc
 label_smoothing: 0.1
 label_smoothing: 0.1
@@ -32,3 +33,6 @@ decoder-ffn-embed-dim: 2048
 decoder-attention-heads: 4
 decoder-attention-heads: 4
 attention-dropout: 0.1
 attention-dropout: 0.1
 activation-dropout: 0.1
 activation-dropout: 0.1
+
+#load-pretrained-encoder-from:
+#load-pretrained-decoder-from:
\ No newline at end of file

egs/libri_trans/asr/conf/conformer.yaml

@@ -2,3 +2,4 @@ macaron-style: True
 use-cnn-module: True
 use-cnn-module: True
 cnn-module-kernel: 31
 cnn-module-kernel: 31
 encoder-attention-type: rel_pos
 encoder-attention-type: rel_pos
+encoder-activation-fn: swish
\ No newline at end of file

egs/libri_trans/asr/conf/debug.yaml

@@ -37,4 +37,4 @@ macaron-style: True
 use-cnn-module: True
 use-cnn-module: True
 cnn-module-kernel: 31
 cnn-module-kernel: 31
 encoder-activation-fn: swish
 encoder-activation-fn: swish
-encoder-attention-type: rel_pos
+encoder-attention-type: rel_pos_legacy

egs/libri_trans/asr/conf/purectc.yaml

@@ -4,22 +4,28 @@ clip-norm: 10.0
 lr-scheduler: inverse_sqrt
 lr-scheduler: inverse_sqrt
 warmup-init-lr: 1e-7
 warmup-init-lr: 1e-7
 warmup-updates: 10000
 warmup-updates: 10000
+weight-decay: 1e-6
 lr: 2e-3
 lr: 2e-3
-#adam_betas: (0.9,0.98)
+adam_betas: (0.9,0.98)
 
 
 criterion: ctc
 criterion: ctc
 zero_infinity: True
 zero_infinity: True
 post-process: sentencepiece
 post-process: sentencepiece
-label_smoothing: 0.1
 
 
-conv-kernel-sizes: 5,5
-conv-channels: 1024
+subsampling-type: conv1d
+subsmapling-layers: 2
+subsampling-filter: 1024
+subsampling-kernel: 5
+subsampling-stride: 2
+subsampling-norm: none
+subsampling-activation: glu
+
 dropout: 0.1
 dropout: 0.1
+attention-dropout: 0.1
+activation-dropout: 0.1
+
 activation-fn: relu
 activation-fn: relu
 encoder-embed-dim: 256
 encoder-embed-dim: 256
 encoder-ffn-embed-dim: 2048
 encoder-ffn-embed-dim: 2048
 encoder-layers: 12
 encoder-layers: 12
 encoder-attention-heads: 4
 encoder-attention-heads: 4
\ No newline at end of file
-
-attention-dropout: 0.1
-activation-dropout: 0.1
\ No newline at end of file

egs/libri_trans/asr/conf/rpr.yaml

-encoder-attention-type: rel_pos
+encoder-attention-type: rel_selfattn
 #encoder-attention-type: relative
 #encoder-attention-type: relative
+#decoder-attention-type: relative
 #max-encoder-relative-length: 100
 #max-encoder-relative-length: 100
+#max-decoder-relative-length: 20

egs/libri_trans/mt/conf/base.yaml

@@ -5,6 +5,7 @@ clip-norm: 10.0
 lr-scheduler: inverse_sqrt
 lr-scheduler: inverse_sqrt
 warmup-init-lr: 1e-7
 warmup-init-lr: 1e-7
 warmup-updates: 8000
 warmup-updates: 8000
+weight-decay: 1e-6
 lr: 1e-3
 lr: 1e-3
 adam_betas: (0.9,0.997)
 adam_betas: (0.9,0.997)
 
 

egs/libri_trans/st/conf/base.yaml

@@ -5,14 +5,21 @@ clip-norm: 10.0
 lr-scheduler: inverse_sqrt
 lr-scheduler: inverse_sqrt
 warmup-init-lr: 1e-7
 warmup-init-lr: 1e-7
 warmup-updates: 10000
 warmup-updates: 10000
+weight-decay: 1e-6
 lr: 2e-3
 lr: 2e-3
-#adam_betas: (0.9,0.98)
+adam_betas: (0.9,0.98)
 
 
 criterion: label_smoothed_cross_entropy_with_ctc
 criterion: label_smoothed_cross_entropy_with_ctc
 label_smoothing: 0.1
 label_smoothing: 0.1
 
 
-conv-kernel-sizes: 5,5
-conv-channels: 1024
+subsampling-type: conv1d
+subsmapling-layers: 2
+subsampling-filter: 1024
+subsampling-kernel: 5
+subsampling-stride: 2
+subsampling-norm: none
+subsampling-activation: glu
+
 dropout: 0.1
 dropout: 0.1
 activation-fn: relu
 activation-fn: relu
 encoder-embed-dim: 256
 encoder-embed-dim: 256

egs/libri_trans/st/conf/conformer.yaml

 macaron-style: True
 macaron-style: True
 use-cnn-module: True
 use-cnn-module: True
 cnn-module-kernel: 31
 cnn-module-kernel: 31
+encoder-attention-type: rel_pos
+encoder-activation-fn: swish
\ No newline at end of file

egs/libri_trans/st/conf/ctc.yaml

 ctc-weight: 0.3
 ctc-weight: 0.3
+zero_infinity: True
+post-process: sentencepiece
\ No newline at end of file

egs/libri_trans/st/conf/sate_ctc.yaml

@@ -14,8 +14,15 @@ label_smoothing: 0.1
 
 
 encoder-normalize-before: True
 encoder-normalize-before: True
 decoder-normalize-before: True
 decoder-normalize-before: True
-conv-kernel-sizes: 5,5
-conv-channels: 1024
+
+subsampling-type: conv1d
+subsmapling-layers: 2
+subsampling-filter: 1024
+subsampling-kernel: 5
+subsampling-stride: 2
+subsampling-norm: none
+subsampling-activation: glu
+
 dropout: 0.1
 dropout: 0.1
 activation-fn: relu
 activation-fn: relu
 encoder-embed-dim: 256
 encoder-embed-dim: 256

egs/librispeech/asr/conf/ConformerCTCSmall.yaml

 arch: s2t_ctc
 arch: s2t_ctc
 optimizer: adam
 optimizer: adam
-clip-norm: 10.0
+#clip-norm: 10.0
 lr-scheduler: inverse_sqrt
 lr-scheduler: inverse_sqrt
 warmup-init-lr: 1e-7
 warmup-init-lr: 1e-7
 warmup-updates: 10000
 warmup-updates: 10000
+weight-decay: 1e-6
 lr: 0.0015
 lr: 0.0015
 adam_betas: (0.9,0.98)
 adam_betas: (0.9,0.98)
 
 
 criterion: ctc
 criterion: ctc
 post-process: sentencepiece
 post-process: sentencepiece
 
 
-subsampling-type: conv1d
+subsampling-type: conv2d
 subsmapling-layers: 2
 subsmapling-layers: 2
-subsampling-filter: 1024
-subsampling-kernel: 5
+subsampling-filter: 176
+subsampling-kernel: 3
 subsampling-stride: 2
 subsampling-stride: 2
-subsampling-norm: none
-subsampling-activation: glu
+subsampling-norm: batch2d
+subsampling-activation: swish
 
 
 dropout: 0.1
 dropout: 0.1
 activation-fn: relu
 activation-fn: relu

fairseq/models/speech_to_text/s2t_ctc.py

@@ -19,6 +19,8 @@ from fairseq.modules import (
     FairseqDropout,
     FairseqDropout,
     LayerNorm,
     LayerNorm,
     PositionalEmbedding,
     PositionalEmbedding,
+    PositionalEncoding,
+    LegacyRelPositionalEncoding,
     RelPositionalEncoding,
     RelPositionalEncoding,
     S2TTransformerEncoderLayer,
     S2TTransformerEncoderLayer,
     DynamicLinearCombination,
     DynamicLinearCombination,
@@ -462,6 +464,10 @@ class S2TCTCEncoder(FairseqEncoder):
             self.embed_positions = RelPositionalEncoding(
             self.embed_positions = RelPositionalEncoding(
                 args.max_source_positions, args.encoder_embed_dim
                 args.max_source_positions, args.encoder_embed_dim
             )
             )
+        elif self.attn_type == "rel_selfattn":
+            self.embed_positions = LegacyRelPositionalEncoding(
+                args.encoder_embed_dim, args.dropout, args.max_source_positions
+            )
         elif self.attn_type == "rope":
         elif self.attn_type == "rope":
             self.embed_positions = None
             self.embed_positions = None
         else:  # Use absolute positional embedding
         else:  # Use absolute positional embedding
@@ -554,7 +560,7 @@ class S2TCTCEncoder(FairseqEncoder):
         # padding and position embedding
         # padding and position embedding
         encoder_padding_mask = lengths_to_padding_mask(input_lengths)
         encoder_padding_mask = lengths_to_padding_mask(input_lengths)
 
 
-        if self.attn_type == "rel_pos":
+        if self.attn_type == "rel_pos" or self.attn_type == "rel_selfattn":
             positions = self.embed_positions(x)
             positions = self.embed_positions(x)
 
 
         elif self.attn_type == "rope":
         elif self.attn_type == "rope":
@@ -566,7 +572,6 @@ class S2TCTCEncoder(FairseqEncoder):
             positions = None
             positions = None
 
 
         x = self.dropout_module(x)
         x = self.dropout_module(x)
-        # positions = self.dropout_module(positions)
 
 
         # add emb into history
         # add emb into history
         if self.history is not None:
         if self.history is not None:
@@ -698,7 +703,7 @@ class CTCDecoder(object):
             model_path=self.lm_model,
             model_path=self.lm_model,
             alpha=self.lm_weight,
             alpha=self.lm_weight,
             beta=0,
             beta=0,
-            cutoff_top_n=40,
+            cutoff_top_n=self.vocab_size,
             cutoff_prob=1.0,
             cutoff_prob=1.0,
             beam_width=self.beam_size,
             beam_width=self.beam_size,
             num_processes=20,
             num_processes=20,

fairseq/models/speech_to_text/s2t_transformer.py

@@ -19,6 +19,7 @@ from fairseq.modules import (
     FairseqDropout,
     FairseqDropout,
     LayerNorm,
     LayerNorm,
     PositionalEmbedding,
     PositionalEmbedding,
+    LegacyRelPositionalEncoding,
     RelPositionalEncoding,
     RelPositionalEncoding,
     S2TTransformerEncoderLayer,
     S2TTransformerEncoderLayer,
     DynamicLinearCombination,
     DynamicLinearCombination,
@@ -133,9 +134,10 @@ class S2TTransformerModel(FairseqEncoderDecoderModel):
                 "reduced",
                 "reduced",
                 "rel_selfattn",
                 "rel_selfattn",
                 "relative",
                 "relative",
+                "rel_pos_legacy",
                 "rel_pos",
                 "rel_pos",
                 "rope",
                 "rope",
-                "abs"
+                "abs",
             ],
             ],
             help="transformer encoder self-attention layer type"
             help="transformer encoder self-attention layer type"
         )
         )
@@ -499,6 +501,10 @@ class S2TTransformerEncoder(FairseqEncoder):
             self.embed_positions = RelPositionalEncoding(
             self.embed_positions = RelPositionalEncoding(
                 args.max_source_positions, args.encoder_embed_dim
                 args.max_source_positions, args.encoder_embed_dim
             )
             )
+        elif self.attn_type in ["rel_selfattn", "rel_pos_legacy"]:
+            self.embed_positions = LegacyRelPositionalEncoding(
+                args.encoder_embed_dim, args.dropout, args.max_source_positions
+            )
         elif self.attn_type == "rope":
         elif self.attn_type == "rope":
             self.embed_positions = None
             self.embed_positions = None
         else:  # Use absolute positional embedding
         else:  # Use absolute positional embedding
@@ -614,7 +620,7 @@ class S2TTransformerEncoder(FairseqEncoder):
         # padding and position embedding
         # padding and position embedding
         encoder_padding_mask = lengths_to_padding_mask(input_lengths)
         encoder_padding_mask = lengths_to_padding_mask(input_lengths)
 
 
-        if self.attn_type == "rel_pos":
+        if self.attn_type in ["rel_pos", "rel_pos_legacy", "rel_selfattn"]:
             positions = self.embed_positions(x)
             positions = self.embed_positions(x)
 
 
         elif self.attn_type == "rope":
         elif self.attn_type == "rope":
@@ -626,7 +632,6 @@ class S2TTransformerEncoder(FairseqEncoder):
             positions = None
             positions = None
 
 
         x = self.dropout_module(x)
         x = self.dropout_module(x)
-        # positions = self.dropout_module(positions)
 
 
         # add emb into history
         # add emb into history
         if self.history is not None:
         if self.history is not None:

fairseq/modules/__init__.py

@@ -44,15 +44,18 @@ from .transpose_last import TransposeLast
 from .unfold import unfold1d
 from .unfold import unfold1d
 from .transformer_layer import TransformerDecoderLayer, TransformerEncoderLayer
 from .transformer_layer import TransformerDecoderLayer, TransformerEncoderLayer
 from .vggblock import VGGBlock
 from .vggblock import VGGBlock
+from .rotary_positional_embedding import RotaryPositionalEmbedding
+from .positional_encoding import (
+    PositionalEncoding,
+    LegacyRelPositionalEncoding,
+    RelPositionalEncoding,
+)
 from .espnet_multihead_attention import (
 from .espnet_multihead_attention import (
     ESPNETMultiHeadedAttention,
     ESPNETMultiHeadedAttention,
     RelPositionMultiHeadedAttention,
     RelPositionMultiHeadedAttention,
+    LegacyRelPositionMultiHeadedAttention,
     RotaryPositionMultiHeadedAttention,
     RotaryPositionMultiHeadedAttention,
 )
 )
-from .rotary_positional_embedding import RotaryPositionalEmbedding
-from .positional_encoding import (
-    RelPositionalEncoding,
-)
 from .convolution import ConvolutionModule
 from .convolution import ConvolutionModule
 from .s2t_transformer_layer import S2TTransformerEncoderLayer
 from .s2t_transformer_layer import S2TTransformerEncoderLayer
 from .pds_layer import PDSTransformerEncoderLayer
 from .pds_layer import PDSTransformerEncoderLayer
@@ -87,7 +90,6 @@ __all__ = [
     "LightweightConv",
     "LightweightConv",
     "LinearizedConvolution",
     "LinearizedConvolution",
     "LocalMultiheadAttention",
     "LocalMultiheadAttention",
-
     "MultiheadAttention",
     "MultiheadAttention",
     "PositionalEmbedding",
     "PositionalEmbedding",
     "PDSTransformerEncoderLayer",
     "PDSTransformerEncoderLayer",
@@ -107,10 +109,13 @@ __all__ = [
     "TransposeLast",
     "TransposeLast",
     "VGGBlock",
     "VGGBlock",
     "unfold1d",
     "unfold1d",
-    "ESPNETMultiheadedAttention",
+    "ESPNETMultiHeadedAttention",
     "PositionalEmbedding",
     "PositionalEmbedding",
     "RelPositionMultiHeadedAttention",
     "RelPositionMultiHeadedAttention",
+    "PositionalEncoding",
+    "LegacyRelPositionalEncoding",
     "RelPositionalEncoding",
     "RelPositionalEncoding",
+    "LegacyRelPositionMultiHeadedAttention",
     "RotaryPositionalEmbedding",
     "RotaryPositionalEmbedding",
     "RotaryPositionMultiHeadedAttention",
     "RotaryPositionMultiHeadedAttention",
 ]
 ]

fairseq/modules/espnet_multihead_attention.py

@@ -156,7 +156,7 @@ class RelPositionMultiHeadedAttention(ESPNETMultiHeadedAttention):
             query: Query tensor T X B X C
             query: Query tensor T X B X C
             key: Key tensor T X B X C
             key: Key tensor T X B X C
             value: Value tensor T X B X C
             value: Value tensor T X B X C
-            pos_emb: Positional embedding tensor B X 2T-1 X C
+            pos_emb: Positional embedding tensor 2T-1 X B(1) X C
             key_padding_mask: Mask tensor T X B
             key_padding_mask: Mask tensor T X B
         Returns:
         Returns:
             torch.Tensor: Output tensor T X B X C.
             torch.Tensor: Output tensor T X B X C.
@@ -196,6 +196,43 @@ class RelPositionMultiHeadedAttention(ESPNETMultiHeadedAttention):
         return scores, None
         return scores, None
 
 
 
 
+class LegacyRelPositionMultiHeadedAttention(RelPositionMultiHeadedAttention):
+    """Multi-Head Attention layer with relative position encoding (old version).
+
+    Details can be found in https://github.com/espnet/espnet/pull/2816.
+
+    Paper: https://arxiv.org/abs/1901.02860
+
+    Args:
+        n_head (int): The number of heads.
+        n_feat (int): The number of features.
+        dropout_rate (float): Dropout rate.
+        zero_triu (bool): Whether to zero the upper triangular part of attention matrix.
+    """
+    def __init__(self, n_feat, n_head, dropout, zero_triu=False):
+        """Construct an RelPositionMultiHeadedAttention object."""
+        super().__init__(n_feat, n_head, dropout, zero_triu)
+
+    def rel_shift(self, x):
+        """Compute relative positional encoding.
+        Args:
+            x: Input tensor B X n_head X T X 2T-1
+        Returns:
+            torch.Tensor: Output tensor.
+        """
+        zero_pad = torch.zeros((*x.size()[:3], 1), device=x.device, dtype=x.dtype)
+        x_padded = torch.cat([zero_pad, x], dim=-1)
+
+        x_padded = x_padded.view(*x.size()[:2], x.size(3) + 1, x.size(2))
+        x = x_padded[:, :, 1:].view_as(x)
+
+        if self.zero_triu:
+            ones = torch.ones((x.size(2), x.size(3)), device=x.device)
+            x = x * torch.tril(ones, x.size(3) - x.size(2))[None, None, :, :]
+
+        return x
+
+
 class RotaryPositionMultiHeadedAttention(ESPNETMultiHeadedAttention):
 class RotaryPositionMultiHeadedAttention(ESPNETMultiHeadedAttention):
     def __init__(
     def __init__(
         self,
         self,

fairseq/modules/positional_encoding.py

@@ -63,13 +63,50 @@ class PositionalEncoding(nn.Module):
         return self.dropout(x)
         return self.dropout(x)
 
 
 
 
+class LegacyRelPositionalEncoding(PositionalEncoding):
+    """Relative positional encoding module (old version).
+
+    Details can be found in https://github.com/espnet/espnet/pull/2816.
+
+    See : Appendix B in https://arxiv.org/abs/1901.02860
+
+    Args:
+        d_model (int): Embedding dimension.
+        dropout_rate (float): Dropout rate.
+        max_len (int): Maximum input length.
+
+    """
+
+    def __init__(self, d_model, dropout_rate, max_len=5000):
+        """Initialize class."""
+        super().__init__(
+            d_model=d_model,
+            dropout_rate=dropout_rate,
+            max_len=max_len,
+            reverse=True,
+        )
+
+    def forward(self, x):
+        """Add positional encoding.
+        Args:
+            x : Input tensor T X B X C.
+        Returns:
+            torch.Tensor: Encoded tensor T X B X C.
+
+        """
+        x = x.transpose(0, 1)  # Change TBC to BTC
+        self.extend_pe(x)
+        pos_emb = self.pe[:, : x.size(1)]
+        pos_emb = pos_emb.transpose(0, 1)  # change to TBC
+        return self.dropout(pos_emb)
+
+
 class RelPositionalEncoding(nn.Module):
 class RelPositionalEncoding(nn.Module):
     """Relative positional encoding module (new implementation).
     """Relative positional encoding module (new implementation).
 
 
     Args:
     Args:
-        d_model: Embedding dimension.
-        dropout_rate: Dropout rate.
         max_len: Maximum input length.
         max_len: Maximum input length.
+        d_model: Embedding dimension.
     """
     """
 
 
     def __init__(self, max_len, d_model):
     def __init__(self, max_len, d_model):

fairseq/modules/rel_position_multihead_attention.py

@@ -247,6 +247,7 @@ class RelPositionMultiheadAttention(MultiheadAttention):
                     dim=1,
                     dim=1,
                 )
                 )
 
 
+        pos_emb = pos_emb.repeat(bsz, 1, 1)
         pos_emb = pos_emb.transpose(0, 1)
         pos_emb = pos_emb.transpose(0, 1)
         p = self.linear_pos(pos_emb).view(bsz, -1, self.num_heads, self.head_dim)
         p = self.linear_pos(pos_emb).view(bsz, -1, self.num_heads, self.head_dim)
         # p (bsz * num_heads, tgt_len, head_dim)
         # p (bsz * num_heads, tgt_len, head_dim)

fairseq/modules/s2t_transformer_layer.py

@@ -15,11 +15,12 @@ from fairseq.modules import (
     ConvolutionModule,
     ConvolutionModule,
     ESPNETMultiHeadedAttention,
     ESPNETMultiHeadedAttention,
     RelPositionMultiHeadedAttention,
     RelPositionMultiHeadedAttention,
+    LegacyRelPositionMultiHeadedAttention,
     RotaryPositionMultiHeadedAttention,
     RotaryPositionMultiHeadedAttention,
 )
 )
 from fairseq.modules.fairseq_dropout import FairseqDropout
 from fairseq.modules.fairseq_dropout import FairseqDropout
 from torch import Tensor
 from torch import Tensor
-from fairseq.modules.activations import get_activation_fn, get_activation_class
+from fairseq.modules.activations import get_activation_class
 
 
 
 
 class FeedForwardModule(torch.nn.Module):
 class FeedForwardModule(torch.nn.Module):
@@ -160,8 +161,8 @@ class S2TTransformerEncoderLayer(nn.Module):
             else:
             else:
                 print("The maximum encoder relative length %d can not be -1!" % max_relative_length)
                 print("The maximum encoder relative length %d can not be -1!" % max_relative_length)
                 exit(1)
                 exit(1)
-        elif self.attn_type == "rel_pos":
-            return RelPositionMultiHeadedAttention(
+        elif self.attn_type in ["rel_pos", "rel_pos_legacy"]:
+            return LegacyRelPositionMultiHeadedAttention(
                 embed_dim,
                 embed_dim,
                 attention_heads,
                 attention_heads,
                 dropout=dropout,
                 dropout=dropout,
@@ -225,7 +226,7 @@ class S2TTransformerEncoderLayer(nn.Module):
                 `attn_mask[tgt_i, src_j] = 1` means that when calculating the
                 `attn_mask[tgt_i, src_j] = 1` means that when calculating the
                 embedding for `tgt_i`, we exclude (mask out) `src_j`. This is
                 embedding for `tgt_i`, we exclude (mask out) `src_j`. This is
                 useful for strided self-attention.
                 useful for strided self-attention.
-            positions (Tensor): the position embedding for relative position encoding
+            pos_emb (Tensor): the position embedding for relative position encoding
 
 
         Returns:
         Returns:
             encoded output of shape `(seq_len, batch, embed_dim)`
             encoded output of shape `(seq_len, batch, embed_dim)`
@@ -251,7 +252,7 @@ class S2TTransformerEncoderLayer(nn.Module):
         residual = x
         residual = x
         if self.normalize_before:
         if self.normalize_before:
             x = self.self_attn_layer_norm(x)
             x = self.self_attn_layer_norm(x)
-        if self.attn_type == "rel_selfattn" or self.attn_type == "rel_pos":
+        if self.attn_type in ["rel_pos", "rel_pos_legacy", "rel_selfattn"]:
             assert pos_emb is not None, "Positions is necessary for RPE!"
             assert pos_emb is not None, "Positions is necessary for RPE!"
             x, _ = self.self_attn(
             x, _ = self.self_attn(
                 query=x,
                 query=x,