Google Sound

Logo
Frontend Webpage

Inspiration

Reading books is important, it doesn't matter if it's a textbook or a novel, but sometimes we don't have the time to do so or we can't sit still and read. So to solve the problem audiobooks were invented but not all books are available in audio files. And so to solve the new problem we are sought to be able to convert any books or articles into audio files with the help of AI.

What it does

We used OCR to extract text from any given image. With the extracted text we made a script to summarize the text in an easier-to-understand way and convert text into audio via TTS services in order to play back any text from an image in a .mp3 format.

How we built it

We used Python for the backend development since it is a primary language used in machine learning and had libraries that were relevant to our project. We used several YouTube tutorials and online guides as well as support from our mentors in order to successfully build our project.

Challenges we ran into

We ran into several challenges with having to learn many concepts in order to fix bugs in our code and make it work well together. Specifically, learning a new language from scratch in one day only using online services and a couple of hours with some mentors.

Accomplishments that we're proud of

We are proud of successfully getting the project to work and having a unique front end since we are all newer to that side of programming. We are also proud of how we did since this is the first hackathon for us.

What we learned

We learned a lot about building websites in the front end with HTML, CSS, and JS. In the backend, we learned how to use AI models to perform text recognition and figured out how to interpret the data in order to get the results we wanted.