LauraGPT
commented
9 months ago 这段代码：
from funasr import AutoModel model = AutoModel( model="damo/speech_seaco_paraformer_large_asr_nat-zh-cn-16k-common-vocab8404-pytorch", model_revision="v2.0.0", 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="v2.0.1", # spk_model="damo/speech_campplus_sv_zh-cn_16k-common", # spk_model_revision="v2.0.0" ) res = model( input="/content/combined_audio_trim.mp3", hotword='磨搭') if __name__ == '__main__': print(res) print(res,'\n',res[0]['text_with_punc'])之前在colab上可以运行，识别效果不错，前两天运行不了了，报错： 2 frames /usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, *kwargs) 1525 or _global_backward_pre_hooks or _global_backward_hooks 1526 or _global_forward_hooks or _global_forward_pre_hooks): -> 1527 return forward_call(args, **kwargs) 1528 1529 try:
TypeError: SeacoParaformer.forward() missing 3 required positional arguments: 'speech_lengths', 'text', and 'text_lengths'
后来换了项目主页上的代码：
from funasr import AutoModel # paraformer-zh is a multi-functional asr model # 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", \ spk_model="cam++", spk_model_revision="v2.0.2") res = model.generate(input=f"/content/combined_audio_trim (9).mp3", batch_size=64, hotword='魔搭') print(res)还是运行不了，报错是： /usr/local/lib/python3.10/dist-packages/torch/_tensor.py in array(self, dtype) 1028 return handle_torch_function(Tensor.array, (self,), self, dtype=dtype) 1029 if dtype is None: -> 1030 return self.numpy() 1031 else: 1032 return self.numpy().astype(dtype, copy=False)
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
请多多指教。
Please note the usage model.generate(input=)
from funasr import AutoModel
# paraformer-zh is a multi-functional asr model
# 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", \
spk_model="cam++", spk_model_revision="v2.0.2")
res = model.generate(input=f"{model.model_path}/example/asr_example.wav",
batch_size=64,
hotword='魔搭')
print(res)
eleven-monkey
dsyrock
这段代码：
之前在colab上可以运行，识别效果不错，前两天运行不了了，报错： 2 frames /usr/local/lib/python3.10/dist-packages/torch/nn/modules/module.py in _call_impl(self, *args, *kwargs) 1525 or _global_backward_pre_hooks or _global_backward_hooks 1526 or _global_forward_hooks or _global_forward_pre_hooks): -> 1527 return forward_call(args, **kwargs) 1528 1529 try:
TypeError: SeacoParaformer.forward() missing 3 required positional arguments: 'speech_lengths', 'text', and 'text_lengths'
后来换了项目主页上的代码：
还是运行不了，报错是： /usr/local/lib/python3.10/dist-packages/torch/_tensor.py in array(self, dtype) 1028 return handle_torch_function(Tensor.array, (self,), self, dtype=dtype) 1029 if dtype is None: -> 1030 return self.numpy() 1031 else: 1032 return self.numpy().astype(dtype, copy=False)
TypeError: can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first.
请多多指教。