我尝试用modelscope-funasr调用paraformer报错，怎么解决？
我尝试用modelscope-funasr调用paraformer报错，怎么解决？from modelscope.pipelines import pipeline
from modelscope.utils.constant import Tasks
inference_pipeline = pipeline(
task=Tasks.auto_speech_recognition,
model='iic/speech_paraformer-large-contextual_asr_nat-zh-cn-16k-common-vocab8404', model_revision="v2.0.4")
报错：OSError: FunASRPipeline: /home/robot/miniconda3/envs/ai/lib/python3.8/site-packages/torchaudio/lib/libtorchaudio.so: undefined symbol: _ZN2at4_ops10zeros_like4callERKNS_6TensorEN3c108optionalINS5_10ScalarTypeEEENS6_INS5_6LayoutEEENS6_INS5_6DeviceEEENS6_IbEENS6_INS5_12MemoryFormatEEE
OSError: /home/robot/miniconda3/envs/ai/lib/python3.8/site-packages/torchaudio/lib/libtorchaudio.so: undefined symbol: _ZN2at4_ops10zeros_like4callERKNS_6TensorEN3c108optionalINS5_10ScalarTypeEEENS6_INS5_6LayoutEEENS6_INS5_6DeviceEEENS6_IbEENS6_INS5_12MemoryFormatEEE
-
orchaudio版本不对，网上查阅一下版本关系，cuda，torch，torchaudio。此回答整理自钉群“modelscope-funasr社区交流”
2024-02-28 16:41:46赞同 展开评论