In the speechbrain/asr-wav2vec2-ctc-aishell task, how can I load a local model for inference instead of downloading it from HuggingFace? #2745

`from speechbrain.inference.interfaces import foreign_class

file_path = "C:/Users/Lenovo/Desktop/ASR_prjs/data_aishell/data_aishell/wav/test/S0764/BAC009S0764W0121.wav"
asr_model = foreign_class(source="C:/Users/Lenovo/Desktop/ASR_prjs/speechbrainasr-wav2vec2-ctc-aishell", pymodule_file="custom_interface.py", classname="CustomEncoderDecoderASR")
print(asr_model.transcribe_file(file_path))`

Traceback (most recent call last):
  File "c:\Users\Lenovo\Desktop\ASR_prjs\speechbrain\debug.py", line 4, in <module>
    asr_model = foreign_class(source="C:/Users/Lenovo/Desktop/ASR_prjs/speechbrainasr-wav2vec2-ctc-aishell", pymodule_file="custom_interface.py", classname="CustomEncoderDecoderASR")
  File "c:\Users\Lenovo\Desktop\ASR_prjs\speechbrain\speechbrain\inference\interfaces.py", line 109, in foreign_class
    hparams_local_path = fetch(
  File "c:\Users\Lenovo\Desktop\ASR_prjs\speechbrain\speechbrain\utils\fetching.py", line 387, in fetch
    fetched_file = huggingface_hub.hf_hub_download(
  File "C:\softwares\Anaconda\path\envs\speech\lib\site-packages\huggingface_hub\utils\_validators.py", line 106, in _inner_fn
    validate_repo_id(arg_value)
  File "C:\softwares\Anaconda\path\envs\speech\lib\site-packages\huggingface_hub\utils\_validators.py", line 154, in validate_repo_id
    raise HFValidationError(
huggingface_hub.errors.HFValidationError: Repo id must be in the form 'repo_name' or 'namespace/repo_name': 'C:/Users/Lenovo/Desktop/ASR_prjs/speechbrainasr-wav2vec2-ctc-aishell'. Use `repo_type` argument if needed.