Reads a NES Tetris video and adds only the Tetris sound effect.

• Select either detection by flashing pixel or by line count OCR (flashing is much faster)
• Use the file dialog to select the video file, and press "Go to pixel selection..."
• Depending on which method you chose, click on the flashing pixel or on the top-left and bottom-right corners of the line count number box

(this is just for illustration, the points ain't drawn on the screen yet)

• If needed you can go forward on the video by pressing 'n', press 'q' when you are done (might have to press a few times for some reason)
• This window will close and another one will appear playing the video and detecting tetrises, when it finishes, it will automatically create a file named "video_with_tetris_sound_effect.mp4" with a copy of the video and the tetris sounds.

The windows executable should be available here (click on on the most recent green ticked link and scroll to the bottom of the page, download requires github login.

If you want to run it from the python file, install tesserocr, download the repository and inside the directory run:

python3 -m venv venv
source venv/bin/activate (on Windows run venv\Scripts\activate.bat)
pip install -r requirements.txt
python tetris_sound.py

Note: FFmpeg has to be on your path in either case as of now.

(I'm not very familiar with python (or openCV and OCR), so the code is a bit of a mess)

This project is under the GPLv2 or later.

The file eng.traineddata was added here for convenience, it is from this repository and is under the Apache-2.0 license.