The ultimate motivation for this application came from a desire for better sleep and, by extension, a better way to fall asleep. The first step was to look at the different things people use to fall asleep. Two commonly used media are audiobooks and podcasts. We noticed that though these are indeed very helpful for falling asleep, they weren't specifically designed for it. This niche that was missing is what CABbAGE was born from. We realized that there was no reason to risk losing your place in a podcast if we could generate the podcast audio on the fly. This method would allow the user to select the "sound" of audio that they want to fall asleep to, without having to find an actual podcast episode to fit it. This application allows a user to gain the proven benefit of falling asleep to a white noise, without the downsides associated with using podcasts or audiobooks that aren't designed for sleep.
See paper: https://github.com/jshom/csci-4964-cabbage/blob/main/paper/cabbage-paper.pdf
Using make to spin up for build system
Use make paper command to render the paper, make sure you have Pandoc installed.
Use make app command to spin up the electron app, you will first need to enter the /app directory and run npm install to fetch dependencies
Use make notebook command to spin up the Jupyter notebook.
You will need to have docker and docker-compose installed.
See https://www.docker.com/ and https://docs.docker.com/compose/ for docs and installation.
To start up run make, this will spin up a Jupyter notebook. All files will be
saved to the ./notebooks directory.
Tensorflow, Keras and Numpy should is pre-installed. The IBM Watson library should aso be installed. Install library using:
pip install --upgrade "ibm-watson>=4.7.1"NOTE: using the nightly tf jupyer notebook image, hoping it doesn't cause any issues
- @EldritchCrow Sam Cohen
- @GregPikitis Emelia Blankenship
- @gizzon Nicole Gizzo
- @IkeLyons Ike Lyons
- @sen-francis Sen Francis
- @jshom Jacob Shomstein