AI OCR

Home Page
AI Enabled OCR
Auto Language Detection
Translate to native language
State of the art language translation
Custom audio generation of translated text

Inspiration

Inspired by a desire to leverage existing industries to help increase African literacy rate and thereby improving the economy as well.

What it does

AI OCR is an AI-enabled web application which is capable of taking an image that contains text of non-African language and converts it to one of Africa's native Languages - Swahili, Arabic, Afrikaans and vice versa.

How I built it

I built this using Python. The libraries I used to build this application include google translate API to convert language from one to another, easyocr for optical character recognition and gTTS library for voice output with a native accent.

Challenges I ran into

Few of the challenges I ran into while building this project include learning new technologies like streamlit for website development and reading open-source documentation as I used a few libraries that are very new and a lot of tutorials are not available online.

Accomplishments that I'm proud of

Being able to bring a fully functional web application that is actually capable of affecting the lives of millions of people for the good and enabling them to achieve new heights with the help of technology.

What I learned

Reading open source documentation. Creating a solution after framing the problem correctly and a general overview of web development as this was my first attempt at building one.

What's next for AI OCR

Scaling the application so that it can be used by a large number of people. I also plan to add more handy features like entity extraction in the near future.