Largely inspired by frustrations in my academic workflows, I wanted a way to easily share explanations/concepts with my friends, just like how I use Snipping Tool.
ClipIt is built ontop of this idea with the goal of making it easy to share videos clips with others.
The screen recording is handled by a python script, clip.py, which uses
OpenCV and listens for a keypress to start/stop the recording. For my use,
this is set as F4 and F8 to start and stop the recording, respectively. Addition:
Microphone support is enabled when you hit F4 and will stop recording audio input
when F8 is pressed. Ffmpeg is used to stitch together the audio and video clip.
This is editable in the clip.py file, found in the root of the project.
The frontend used Next.js, as it this was also a learning opportunity for me to try out new JS frameworks.
One of the major design decisions was to minimize the footprint and make it intuitive as Snipping Tool or Lightshot. The idea is to serve these video clips via a link through an ngrok tunnel and send the link with the latest clip.
pip install -r requirements.txt
pip install pipwin
pipwin install pyaudio
cd clipit
npm install
python clip.py // In root
cd clipit
npm run dev
npm run ngrok
One particular challenge I faced was running functions concurrently. It turns out that python Thread or Multiprocessing libraries have a slight delay in the execution of the functions (recording Audio and Video), which doesn't meet my UX requirements. I found that I had to press F8 twice inorder to end the Audio recording. However, as this comment suggested, the Ray library is able to run both the Audio and Video recording functions at the same time.
I'm always open to improving the quality of this project. If there's a feature request, performance improvement, or bug fixes that you found, feel free to submit a PR!