AssertionError: cam++ is not registered on CUDA12.1

[ozora-ogino]

Notice: In order to resolve issues more efficiently, please raise issue following the template. （注意：为了更加高效率解决您遇到的问题，请按照模板提问，补充细节）

🐛 Bug

To Reproduce

Steps to reproduce the behavior (always include the command you ran):

environment: cuda:12.1, torch:2.3.1, funasr:1.0.30, modelscope: 1.15.0

Run the code sample on this dockerfile

ARG PYTORCH="2.3.1"
ARG CUDA="12.1"
ARG CUDNN="8"

FROM pytorch/pytorch:${PYTORCH}-cuda${CUDA}-cudnn${CUDNN}-devel

ENV TORCH_CUDA_ARCH_LIST="6.0 6.1 7.0+PTX"
ENV TORCH_NVCC_FLAGS="-Xfatbin -compress-all"
ENV CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"

RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/3bf863cc.pub
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/machine-learning/repos/ubuntu1804/x86_64/7fa2af80.pub

RUN apt-get update && apt-get install -y ffmpeg libsm6 libxext6 git ninja-build libglib2.0-0 libsm6 libxrender-dev libxext6 \
	&& apt-get clean \
	&& rm -rf /var/lib/apt/lists/*

RUN pip install --upgrade pip && \
	pip install --no-cache-dir librosa soundfile faster-whisper "numpy<2.0.0"

RUN pip install openai google-cloud-speech wandb pydub bert_score evaluate jiwer funasr modelscope

WORKDIR /workspace

Code sample

When I run following code error happen on cuda:12.1 but not happend when I use cuda 11.6.

from funasr import AutoModel
model = AutoModel(model="cam++", model_revision="v2.0.2")

Error message:

2024-07-03 05:06:22,007 - modelscope - INFO - PyTorch version 2.3.1 Found.
2024-07-03 05:06:22,008 - modelscope - INFO - Loading ast index from /root/.cache/modelscope/ast_indexer
2024-07-03 05:06:22,044 - modelscope - INFO - Updating the files for the changes of local files, first time updating will take longer time! Please wait till updating done!
2024-07-03 05:06:22,058 - modelscope - INFO - AST-Scanning the path "/opt/conda/lib/python3.10/site-packages/modelscope" with the following sub folders ['models', 'metrics', 'pipelines', 'preprocessors', 'trainers', 'msdatasets', 'exporters']
Download: iic/speech_campplus_sv_zh-cn_16k-common failed!: During ast indexing the file /opt/conda/lib/python3.10/site-packages/modelscope/models/cv/indoor_layout_estimation/networks/utils.py, a related error excepted in the file /opt/conda/lib/python3.10/site-packages/modelscope/utils/ast_utils.py at line: 152: "attr = getattr(node, field)" with error msg: "AttributeError: 'ClassDef' object has no attribute 'type_params'", please double check the origin file /opt/conda/lib/python3.10/site-packages/modelscope/models/cv/indoor_layout_estimation/networks/utils.py to see whether the file is correctly edited.
Traceback (most recent call last):
  File "/workspace/src/diarizer.py", line 112, in <module>
    diarize(wav_files, dest)
  File "/workspace/src/diarizer.py", line 84, in diarize
    embed_model = CamPlus()
  File "/workspace/src/diarizer.py", line 45, in __init__
    self.model = AutoModel(model="cam++", model_revision="v2.0.2")
  File "/opt/conda/lib/python3.10/site-packages/funasr/auto/auto_model.py", line 124, in __init__
    model, kwargs = self.build_model(**kwargs)
  File "/opt/conda/lib/python3.10/site-packages/funasr/auto/auto_model.py", line 218, in build_model
    assert model_class is not None, f'{kwargs["model"]} is not registered'
AssertionError: cam++ is not registered

Expected behavior

Cam++ model is loaded without error.

Environment

• OS (e.g., Linux): Linux
• FunASR Version (e.g., 1.0.0): 1.0.30
• ModelScope Version (e.g., 1.11.0): 1.15.0
• PyTorch Version (e.g., 2.0.0): 2.3.1
• How you installed funasr (pip, source): pip
• Python version: 3.10
• GPU (e.g., V100M32)
• CUDA/cuDNN version (e.g., cuda11.7): cuda12.1
• Docker version (e.g., funasr-runtime-sdk-cpu-0.4.1)
• Any other relevant information:

Additional context