About

This tool extracts clips from multiple video files, resizes them to the same resolution, and combines them into a single "recap" video file, with crossfade and audio normalisation, as well as custom subtitles. It is compatible with Linux, Mac, and Windows.

The user can either specify their desired clips manually or let recap-video-generator automatically select the best clip for each video by detecting choruses within the audio.

Requirements

This tool uses the third party Python modules OpenPyXL, MoviePy, and Pychorus. Furthermore, MoviePy requires ImageMagick. Please install these before attempting to run the Python scripts included here. Note that MoviePy uses features of Pillow 9.5.0 that are no longer supported in the most recent version of Pillow, so you may need to roll back your version of Pillow. Do this by typing pip install Pillow==9.5.0 in the terminal.

If you are on Linux/Mac, you will need to edit ImageMagick's access permissions in policy.xml in order for MoviePy to work properly:

1. Type identify -list policy in the terminal. The first line should tell you where your policy.xml file is located.
2. Type sudo nano /etc/ImageMagick-6/policy.xml in the terminal, replacing /etc/ImageMagick-6/policy.xml with the address of your policy.xml file.
3. Directly edit the file in the terminal, removing the line that reads <policy domain="path" rights="none" pattern="@*" />. This line should be near the end of the file.
4. Press Ctrl+O to save the changes, then Ctrl+X to exit the editor.

If you are on Windows, you will need to specify the path to ImageMagick in MoviePy's configuration settings:

1. Find magick.exe on your computer. Right-click the file and click 'Copy as path'.
2. Find the config_defaults.py file in MoviePy's folder and open it in Notepad or the editor of your choice.
3. Find the line that says IMAGEMAGICK_BINARY = os.getenv('IMAGEMAGICK_BINARY', 'auto-detect') and change it to IMAGEMAGICK_BINARY = r"C:\Program Files\ImageMagick-7.1.1-Q16-HDRI\magick.exe", where the part in quotation marks should be the path you just copied.
4. Save the changes.

Instructions

1. Put the video files that are to be used for the recap in the Videos folder provided.
2. Enter the following data into video_data.xlsx:
   • The filenames of the video files in the Videos folder, including extensions;
   • (Optional) The start and end times of the video clips you wish to include in the recap, in the form hh:mm:ss;
   • (Optional) Three lines of subtitles for each clip.
3. (Optional) If you wish to use a custom image overlay on your first clip, place your image in the same directory as video_data.xlsx and name it intro.png.
4. Run recap_generator.py and follow the on-screen instructions.
   • Note: If you specify that you are using your first clip as an intro clip, any subtitles for that clip will automatically be enlarged and placed in the centre of the screen. This may be used as an alternative to a custom image overlay.

Note, if the user does not specify a start and end time for a given clip, recap-video-generator will attempt to detect a chorus within the audio and select a clip automatically.