search

modelscope / 3D-Speaker

A Repository for Single- and Multi-modal Speaker Verification, Speaker Recognition and Speaker Diarization
Apache License 2.0
1.23k stars 104 forks source link

在language identification中funasr 的问题 #150

Closed panxin801 closed 1 week ago

panxin801 commented 1 week ago

您好，感谢公开这么好的工作，不过我使用时遇到个问题。 在运行 \egs\3dspeaker\language-identification\run_paraformer.sh 的stage 4 的时候，爆出了

Traceback (most recent call last):
  File "speakerlab/bin/train_para.py", line 25, in <module>
    from funasr.build_utils.build_model_from_file import build_model_from_file
ModuleNotFoundError: No module named 'funasr.build_utils'

modelscope==1.19.2 funasr==1.1.14 Python==3.8.20 请问要怎么解决呢

yfchenlucky commented 1 week ago

先尝试使用funasr==0.8.4，后续有时间我会更新代码。

panxin801 commented 1 week ago

我刚才现在换到了0.8.8 暂时解决了