Merge pull request #355 from alibaba-damo-academy/dev_dzh

add authority

egs_modelscope/speaker_diarization/speech_diarization_sond-en-us-callhome-8k-n16k4-pytorch/infer.py

@@ -1,3 +1,9 @@
+"""
+Author: Speech Lab, Alibaba Group, China
+TOLD: A Novel Two-Stage Overlap-Aware Framework for Speaker Diarization
+https://arxiv.org/abs/2303.05397
+"""
+
 from modelscope.pipelines import pipeline
 from modelscope.utils.constant import Tasks
 

egs_modelscope/speaker_diarization/speech_diarization_sond-zh-cn-alimeeting-16k-n16k4-pytorch/infer.py

@@ -1,3 +1,9 @@
+"""
+Author: Speech Lab, Alibaba Group, China
+SOND: Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
+https://arxiv.org/abs/2211.10243
+"""
+
 from modelscope.pipelines import pipeline
 from modelscope.utils.constant import Tasks
 

funasr/models/e2e_diar_sond.py

@@ -36,8 +36,12 @@ else:
 
 
 class DiarSondModel(AbsESPnetModel):
-    """Speaker overlap-aware neural diarization model
-    reference: https://arxiv.org/abs/2211.10243
+    """
+    Author: Speech Lab, Alibaba Group, China
+    SOND: Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
+    https://arxiv.org/abs/2211.10243
+    TOLD: A Novel Two-Stage Overlap-Aware Framework for Speaker Diarization
+    https://arxiv.org/abs/2303.05397
     """
 
     def __init__(

funasr/models/e2e_sv.py

@@ -1,3 +1,7 @@
+"""
+Author: Speech Lab, Alibaba Group, China
+"""
+
 import logging
 from contextlib import contextmanager
 from distutils.version import LooseVersion

funasr/models/encoder/resnet34_encoder.py

@@ -406,6 +406,12 @@ class ResNet34Diar(ResNet34):
             tf2torch_tensor_name_prefix_torch="encoder",
             tf2torch_tensor_name_prefix_tf="seq2seq/speech_encoder"
     ):
+        """
+        Author: Speech Lab, Alibaba Group, China
+        SOND: Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
+        https://arxiv.org/abs/2211.10243
+        """
+
         super(ResNet34Diar, self).__init__(
             input_size,
             use_head_conv=use_head_conv,
@@ -633,6 +639,12 @@ class ResNet34SpL2RegDiar(ResNet34_SP_L2Reg):
             tf2torch_tensor_name_prefix_torch="encoder",
             tf2torch_tensor_name_prefix_tf="seq2seq/speech_encoder"
     ):
+        """
+        Author: Speech Lab, Alibaba Group, China
+        TOLD: A Novel Two-Stage Overlap-Aware Framework for Speaker Diarization
+        https://arxiv.org/abs/2303.05397
+        """
+
         super(ResNet34SpL2RegDiar, self).__init__(
             input_size,
             use_head_conv=use_head_conv,

funasr/tasks/diar.py

@@ -1,3 +1,11 @@
+"""
+Author: Speech Lab, Alibaba Group, China
+SOND: Speaker Overlap-aware Neural Diarization for Multi-party Meeting Analysis
+https://arxiv.org/abs/2211.10243
+TOLD: A Novel Two-Stage Overlap-Aware Framework for Speaker Diarization
+https://arxiv.org/abs/2303.05397
+"""
+
 import argparse
 import logging
 import os

funasr/tasks/sv.py

@@ -1,3 +1,7 @@
+"""
+Author: Speech Lab, Alibaba Group, China
+"""
+
 import argparse
 import logging
 import os