Detailed Description: 0:00 Biggest Challenges: 1:52 Software Demo: 2:25 Potential Impacts: 4:07 Next Steps: 4:33
We were impressed by how robust and useful Assembly AI was, so we wanted to see what we could do to make it better. We've experienced first hand how harmful audio artifacts can be in audio transcripts, so our project aimed to solve that problem.
What it does
Use a grammar API to find and fix issues in Assembly AI transcriptions, thereby filling in gaps created by audio artifacts.
How we built it
We created a flask website with a python backend and html/css frontend. We used both the Assembly AI API, as well as the TextGears grammar API.
Challenges we ran into
Once we had come up with our idea, our biggest challenges were finding a quality grammar API, as well as developing the frontend. This was our first time working with html/css, so there was a sharp learning curve, but we were able to overcome it.
Accomplishments that we're proud of
One accomplishment is definitely coming up with the idea. Assembly AI is already packed with features, so finding a new one to implement felt great. Also, seeing our hard work on the website's frontend was also really gratifying.
What we learned
We learned a lot about full-stack development, specifically how to lace together frontend and backend.
What's next for AstroChecker
Our next steps are to scale up and create a public utility rather than just having a local demo.