Merge branch 'master' into multi_ja_en_mls_english_clean

egs/aishell/ASR/conformer_ctc/transformer.py

@@ -545,6 +545,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/aishell/ASR/conformer_mmi/transformer.py

@@ -545,6 +545,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/gigaspeech/ASR/conformer_ctc/transformer.py

@@ -549,6 +549,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/librispeech/ASR/conformer_ctc/transformer.py

@@ -549,6 +549,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/librispeech/ASR/conformer_ctc2/transformer.py

@@ -550,6 +550,7 @@ class TransformerDecoderLayer(nn.Module):
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
         warmup: float = 1.0,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/librispeech/ASR/conformer_mmi/transformer.py

@@ -537,6 +537,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/librispeech/ASR/streaming_conformer_ctc/transformer.py

@@ -567,6 +567,7 @@ class TransformerDecoderLayer(nn.Module):
         memory_mask: Optional[torch.Tensor] = None,
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

egs/tedlium3/ASR/conformer_ctc2/transformer.py

@@ -612,6 +612,7 @@ class TransformerDecoderLayer(nn.Module):
         tgt_key_padding_mask: Optional[torch.Tensor] = None,
         memory_key_padding_mask: Optional[torch.Tensor] = None,
         warmup: float = 1.0,
+        **kwargs,
     ) -> torch.Tensor:
         """Pass the inputs (and mask) through the decoder layer.
 

icefall/mmi.py

@@ -124,6 +124,7 @@ def _compute_mmi_loss_exact_non_optimized(
     den_tot_scores = den_lats.get_tot_scores(log_semiring=True, use_double_scores=True)
 
     tot_scores = num_tot_scores - den_scale * den_tot_scores
+    tot_scores = tot_scores.masked_fill(torch.isinf(tot_scores), 0.0)
 
     loss = -1 * tot_scores.sum()
     return loss

icefall/utils.py

@@ -1391,13 +1391,20 @@ def add_eos(ragged: k2.RaggedTensor, eos_id: int) -> k2.RaggedTensor:
     return concat(ragged, eos_id, direction="right")
 
 
-def make_pad_mask(lengths: torch.Tensor, max_len: int = 0) -> torch.Tensor:
+def make_pad_mask(
+    lengths: torch.Tensor,
+    max_len: int = 0,
+    pad_left: bool = False,
+) -> torch.Tensor:
     """
     Args:
       lengths:
         A 1-D tensor containing sentence lengths.
       max_len:
         The length of masks.
+      pad_left:
+        If ``False`` (default), padding is on the right.
+        If ``True``, padding is on the left.
     Returns:
       Return a 2-D bool tensor, where masked positions
       are filled with `True` and non-masked positions are
@@ -1414,9 +1421,14 @@ def make_pad_mask(lengths: torch.Tensor, max_len: int = 0) -> torch.Tensor:
     max_len = max(max_len, lengths.max())
     n = lengths.size(0)
     seq_range = torch.arange(0, max_len, device=lengths.device)
-    expaned_lengths = seq_range.unsqueeze(0).expand(n, max_len)
+    expanded_lengths = seq_range.unsqueeze(0).expand(n, max_len)
 
-    return expaned_lengths >= lengths.unsqueeze(-1)
+    if pad_left:
+        mask = expanded_lengths < (max_len - lengths).unsqueeze(1)
+    else:
+        mask = expanded_lengths >= lengths.unsqueeze(-1)
+
+    return mask
 
 
 # Copied and modified from https://github.com/wenet-e2e/wenet/blob/main/wenet/utils/mask.py