funasr1.0 update (#1278)

README.md

@@ -93,7 +93,7 @@ from funasr import AutoModel
 # use vad, punc, spk or not as you need
 model = AutoModel(model="paraformer-zh", model_revision="v2.0.2",
                   vad_model="fsmn-vad", vad_model_revision="v2.0.2",
-                  punc_model="ct-punc-c", punc_model_revision="v2.0.2",
+                  punc_model="ct-punc-c", punc_model_revision="v2.0.3",
                   # spk_model="cam++", spk_model_revision="v2.0.2",
                   )
 res = model.generate(input=f"{model.model_path}/example/asr_example.wav", 

README_zh.md

@@ -89,7 +89,7 @@ from funasr import AutoModel
 # use vad, punc, spk or not as you need
 model = AutoModel(model="paraformer-zh", model_revision="v2.0.2",
                   vad_model="fsmn-vad", vad_model_revision="v2.0.2",
-                  punc_model="ct-punc-c", punc_model_revision="v2.0.2",
+                  punc_model="ct-punc-c", punc_model_revision="v2.0.3",
                   # spk_model="cam++", spk_model_revision="v2.0.2",
                   )
 res = model.generate(input=f"{model.model_path}/example/asr_example.wav", 

examples/industrial_data_pretraining/bicif_paraformer/demo.py

@@ -10,7 +10,7 @@ model = AutoModel(model="damo/speech_paraformer-large-vad-punc_asr_nat-zh-cn-16k
                   vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch",
                   vad_model_revision="v2.0.2",
                   punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  punc_model_revision="v2.0.2",
+                  punc_model_revision="v2.0.3",
                   spk_model="damo/speech_campplus_sv_zh-cn_16k-common",
                   spk_model_revision="v2.0.2",
                   )

examples/industrial_data_pretraining/bicif_paraformer/infer.sh

@@ -4,7 +4,7 @@ model_revision="v2.0.2"
 vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch"
 vad_model_revision="v2.0.2"
 punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-punc_model_revision="v2.0.2"
+punc_model_revision="v2.0.3"
 spk_model="damo/speech_campplus_sv_zh-cn_16k-common"
 spk_model_revision="v2.0.2"
 

examples/industrial_data_pretraining/paraformer-zh-spk/demo.py

@@ -10,7 +10,7 @@ model = AutoModel(model="damo/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-co
                   vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch",
                   vad_model_revision="v2.0.2",
                   punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  punc_model_revision="v2.0.2",
+                  punc_model_revision="v2.0.3",
                   spk_model="damo/speech_campplus_sv_zh-cn_16k-common",
                   spk_model_revision="v2.0.2"
                   )

examples/industrial_data_pretraining/paraformer-zh-spk/infer.sh

@@ -4,7 +4,7 @@ model_revision="v2.0.2"
 vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch"
 vad_model_revision="v2.0.2"
 punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-punc_model_revision="v2.0.2"
+punc_model_revision="v2.0.3"
 spk_model="damo/speech_campplus_sv_zh-cn_16k-common"
 spk_model_revision="v2.0.2"
 

examples/industrial_data_pretraining/paraformer/demo.py

@@ -5,7 +5,12 @@
 
 from funasr import AutoModel
 
-model = AutoModel(model="damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="v2.0.2")
+model = AutoModel(model="damo/speech_paraformer-large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="v2.0.3",
+                  # vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch",
+                  # vad_model_revision="v2.0.2",
+                  # punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
+                  # punc_model_revision="v2.0.3",
+                  )
 
 res = model.generate(input="https://isv-data.oss-cn-hangzhou.aliyuncs.com/ics/MaaS/ASR/test_audio/asr_example_zh.wav")
 print(res)

examples/industrial_data_pretraining/seaco_paraformer/demo.py

@@ -10,7 +10,7 @@ model = AutoModel(model="damo/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-co
                   vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch",
                   vad_model_revision="v2.0.2",
                   punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch",
-                  punc_model_revision="v2.0.2",
+                  punc_model_revision="v2.0.3",
                   spk_model="damo/speech_campplus_sv_zh-cn_16k-common",
                   spk_model_revision="v2.0.2",
                   )

examples/industrial_data_pretraining/seaco_paraformer/infer.sh

@@ -4,7 +4,7 @@ model_revision="v2.0.2"
 vad_model="damo/speech_fsmn_vad_zh-cn-16k-common-pytorch"
 vad_model_revision="v2.0.2"
 punc_model="damo/punc_ct-transformer_zh-cn-common-vocab272727-pytorch"
-punc_model_revision="v2.0.2"
+punc_model_revision="v2.0.3"
 
 python funasr/bin/inference.py \
 +model=${model} \

funasr/auto/auto_model.py

@@ -391,7 +391,7 @@ class AutoModel:
             if self.punc_model is not None:
                 self.punc_kwargs.update(cfg)
                 punc_res = self.inference(result["text"], model=self.punc_model, kwargs=self.punc_kwargs, **cfg)
-                result["text_with_punc"] = punc_res[0]["text"]
+                result["text"] = punc_res[0]["text"]
                      
             # speaker embedding cluster after resorted
             if self.spk_model is not None:

funasr/models/paraformer/model.py

@@ -451,7 +451,7 @@ class Paraformer(torch.nn.Module):
             self.nbest = kwargs.get("nbest", 1)
         
         meta_data = {}
-        if isinstance(data_in, torch.Tensor): # fbank
+        if isinstance(data_in, torch.Tensor) and kwargs.get("data_type", "sound") == "fbank": # fbank
             speech, speech_lengths = data_in, data_lengths
             if len(speech.shape) < 3:
                 speech = speech[None, :, :]