search

Huanshere / VideoLingo

Netflix-level subtitle cutting, translation, alignment, and even dubbing - one-click fully automated AI video subtitle team | Netflix级字幕切割、翻译、对齐、甚至加上配音，一键全自动视频搬运AI字幕组
https://videolingo.io
Apache License 2.0
4.48k stars 470 forks source link

上传完视频点开始按钮过了几十分钟之后报错connection error 504 #50

Closed wick233 closed 1 month ago

wick233 commented 1 month ago

Connection error Connection failed with status 504, and response " 错误: 不能获取请求的 URL

错误

您所请求的网址（URL）无法获取

当尝试取回该 URL 时遇到下面的错误：http://64.69.41.28:8501/_stcore/health

连接到 64.69.41.28 失败。

系统返回以下内容：(111) Connection refused

The remote host or network may be down. Please try the request again.

缓存服务器的管理员 cncs-netapp@foxconn.com.


".

wick233 commented 1 month ago

又试了几次报错没有找到模块 ModuleNotFoundError: No module named 'whisperx' Traceback: File "/opt/conda/lib/python3.10/site-packages/streamlit/runtime/scriptrunner/exec_code.py", line 88, in exec_func_with_error_handling result = func() File "/opt/conda/lib/python3.10/site-packages/streamlit/runtime/scriptrunner/script_runner.py", line 590, in code_to_exec exec(code, module.dict) File "/app/st.py", line 123, in main() File "/app/st.py", line 117, in main text_processing_section() File "/app/st.py", line 32, in text_processing_section process_text() File "/app/st.py", line 51, in process_text step2_whisper.transcribe(video_file) File "/app/core/step2_whisper.py", line 17, in transcribe from core.all_whisper_methods.whisperX import transcribe as ts File "/app/core/all_whisper_methods/whisperX.py", line 3, in import whisperx

Huanshere commented 1 month ago

是挂在了linux服务器上吗？这个看起来是文件下载回传的网络问题，可能是网络原因也可能是streamlit的原因

wick233 commented 1 month ago

原因也可能是streamlit的原因

用docker部署到了linux上，切换Whisper Method还是会报找不到模块错误，用的是cc洛杉矶的小鸡

Huanshere commented 1 month ago

No module named 'whisperx' 是因为没有安装 third party 文件夹里面的 whisperX 项目，docker 稳定性未知，最好参照本地部署指南进行命令行安装。另外，Linux 部署非 whisper api 版没有测试过，理论上应该没问题～

Huanshere commented 1 month ago

还是推荐使用一键整合包哈哈哈

Huanshere commented 1 month ago

似乎远程传文件容易导致这个问题，这似乎是 streamlit 处理大文件时的弊端，默认上传文件的大小限制在 200M，是我手动改大了，因此还是建议本地上传，或者部署后用链接下载的方式。