Grasp

Grasp is an easy-to-use tool that streamlines the flow of information involved in studying.

Grasp is written primarily for remote-learning students with access to recorded lectures. In addition to aiding students taking notes, it can also be used by educators to provide more resources to students easily.

Grasp takes a recorded video and extracts slides from the video. In addition to extracting slides, it adds references to topics referenced in the slide to the slide's comments. It also exports a summary view of topics in a Markdown formatted text file.

OS Build Tested: Ubuntu LTS 20.04.04 Dependencies: python3.9, pipenv

File Structure

• src/ has all the source code. The main entrypoint is main.py.
• src/tests/ is where all the tests should be located

Running

Python 3.9 should be pre-installed in Ubuntu 20.04. If that is not the case, follow this tutorial: https://linuxize.com/post/how-to-install-python-3-9-on-ubuntu-20-04/

Installing with a script:

1. Install python3.9 and pip3 with apt install python3.9 python3-pip
2. In the project directory, run chmod +x install.sh
3. Run ./install.sh

Manual Installation:

1. Make sure you have python 3.9 installed. Then, run python3.9 -m pip install pipenv.
2. In the project directory, run python3.9 -m pipenv install and python3.9 -m pipenv install --system.
3. Run pre-commit install. This should be enough so, to verify, run pre-commit run --all-files. All processes should pass.
4. Run python3.9 -m pipenv run spacy download en_core_web_sm.
5. Create a .env with touch .env for holding cloud credentials. Write the credentials location with echo ‘GOOGLE_APPLICATION_CREDENTIALS="cloud_credentials.json"’ > .env
6. Fill in cloud_credentials.json with the API key for Google Cloud. This is obtained from Google Cloud. For this project we provide the json file in Appendix 1.

Running Grasp:

cd to the parent directory of grasp. ls should display the directory src. Run pipenv shell to enter the python virtual environment. While in the virtual environment, run python src/main.py to enter the grasp GUI. From there, select a video to parse. Alternatively, pipenv run python src/main.py can be run without entering the pipenv shell.

Troubleshooting:

If running step 3 returns an error, some packages may need to be updated. Update the virtual environment packages with pipenv update. If step 1 still fails, update spacy and black directly. Run pipenv install spacy and pipenv install black. Updating spacy’s dictionary can be performed by pipenv run python -m spacy download en_core_web_sm.