Source Code Deployment on Windows 10/11

0. Go to https://www.python.org/downloads/ to download Python 3.10 for Windows. After downloading, double-click the installer and proceed by clicking "Next." Make sure to check "Add to PATH."

   Open a command prompt and run python -V. If the output is not 3.10, it indicates an installation error or failure to add Python to PATH. Please reinstall.

1. Go to https://github.com/git-for-windows/git/releases/ to download Git. After downloading, double-click the installer and follow the setup steps.

2. Navigate to a folder without spaces or Chinese characters. Type cmd in the address bar and press Enter to open a terminal. Execute all the following commands in this terminal.

3. Run the command: git clone https://github.com/jianchang512/pyvideotrans

4. Continue with: cd pyvideotrans

5. Then run: python -m venv venv

6. Next, execute: .\venv\scripts\activate. After running, check that the command prompt now starts with (venv). If not, an error has occurred.

7. Run pip install -r requirements.txt. If it fails, execute the following two commands to switch the pip mirror to Alibaba's mirror:

pip config set global.index-url https://mirrors.aliyun.com/pypi/simple/
pip config set install.trusted-host mirrors.aliyun.com

Then retry. If it still fails after switching to the Alibaba mirror, try running pip install -r requirements.txt.

8. To enable CUDA acceleration, run the following commands separately:

pip uninstall -y torch torchaudio

pip install torch torchaudio --index-url https://download.pytorch.org/whl/cu126

9. On Windows, to use CUDA acceleration, you must have an NVIDIA GPU and a configured CUDA 12.x environment. For installation details, see CUDA Acceleration Support.

10. Extract ffmpeg.zip into the current source code directory. Overwrite if prompted. After extraction, ensure that ffmpeg.exe and ffprobe.exe are visible in the ffmpeg folder under the source code.

11. Run python sp.py to open the software interface.

Source Code Deployment Issues

1. By default, ctranslate2 version 4.x is used, which only supports CUDA 12.x. If your CUDA version is below 12 and you cannot upgrade to 12.x, run the following commands to uninstall ctranslate2 and reinstall an older version:

pip uninstall -y ctranslate2

pip install ctranslate2==3.24.0

2. If you encounter errors like xx module not found, open requirements.txt, search for the xx module, and remove the == and the version number after it.

3. For a detailed tutorial on installing Python 3.10 on Windows, see