您好,在mac安装部署后输入python3 sp.py,打开并停留在蓝色圆圈初始..

*.158.186.237 /
您好,在mac安装部署后输入python3 sp.py,打开并停留在蓝色圆圈初始页面,右下角显示pyvideotrans V2 loading……,但始终加载不进去,请问怎么正常加载呢?
回复
  • #1 / **.158.90.225
    看看控制台输出信息,是否有报错
  • #2 / **.158.186.237
    Traceback (most recent call last):
    File \"/Users/panda/pyvideotrans-main/sp.py\", line 61, in run
    from videotrans.mainwin.spwin import MainWindow
    File \"/Users/panda/pyvideotrans-main/videotrans/mainwin/spwin.py\", line 22, in
    from videotrans.box import win
    File \"/Users/panda/pyvideotrans-main/videotrans/box/win.py\", line 16, in
    from videotrans.box.worker import Worker, WorkerWhisper, WorkerTTS, FanyiWorker
    File \"/Users/panda/pyvideotrans-main/videotrans/box/worker.py\", line 15, in
    from videotrans.recognition import run as run_recogn
    File \"/Users/panda/pyvideotrans-main/videotrans/recognition/__init__.py\", line 9, in
    from .openai import recogn as openai_recogn
    File \"/Users/panda/pyvideotrans-main/videotrans/recognition/openai.py\", line 14, in
    from whisper.utils import get_writer
    ModuleNotFoundError: No module named \'whisper.utils\'; \'whisper\' is not a package
    重装了几遍whisper还是提示报错
  • #3 / **.158.90.203
    缺少模块啊
    重新安装吧,确定是在同一个虚拟环境下
    再单独安装下
    pip3 install openai-whisper
  • #4 / **.70.223.166
    感谢解答,成功运行了!不过感觉视频分离音频识别字幕还可以加强,用自己翻译的一个视频实验,本来有500多句对话,但用faster large-3模型只识别出了77句,并且当切换其他模型重新识别时,每次的字幕结果都和第一次相同。不知道有没有改进的可能。谢谢!
  • #5 / **.70.211.206
    1. 换更大模型
    2. 选择 清空缓存
问答模块基于php8.1/sqlite/webman/layui构建 © pyvideotrans.com