Initial commit: FunASR Speech Recognition Toolkit

Add complete FunASR codebase including models, runtime, and documentation.

funasr/models/ct_transformer_streaming/attention.py

@@ -0,0 +1,34 @@
+#!/usr/bin/env python3
+# -*- encoding: utf-8 -*-
+# Copyright FunASR (https://github.com/alibaba-damo-academy/FunASR). All Rights Reserved.
+#  MIT License  (https://opensource.org/licenses/MIT)
+
+import torch
+from funasr.models.sanm.attention import MultiHeadedAttentionSANM
+
+
+class MultiHeadedAttentionSANMwithMask(MultiHeadedAttentionSANM):
+    def __init__(self, *args, **kwargs):
+        """Initialize MultiHeadedAttentionSANMwithMask.
+        
+            Args:
+                *args: Variable positional arguments.
+                **kwargs: Additional keyword arguments.
+            """
+        super().__init__(*args, **kwargs)
+
+    def forward(self, x, mask, mask_shfit_chunk=None, mask_att_chunk_encoder=None):
+        """Forward pass for training.
+        
+            Args:
+                x: TODO.
+                mask: TODO.
+                mask_shfit_chunk: TODO.
+                mask_att_chunk_encoder: TODO.
+            """
+        q_h, k_h, v_h, v = self.forward_qkv(x)
+        fsmn_memory = self.forward_fsmn(v, mask[0], mask_shfit_chunk)
+        q_h = q_h * self.d_k ** (-0.5)
+        scores = torch.matmul(q_h, k_h.transpose(-2, -1))
+        att_outs = self.forward_attention(v_h, scores, mask[1], mask_att_chunk_encoder)
+        return att_outs + fsmn_memory