EyeLight

Inspiration

The inspiration for the EyeLight project stems from the need to convert the visual text to audible audio

How we built it

EyeLight is built using:

Streamlit: This Python library simplifies the creation of web applications with minimal effort, enabling developers to design interactive and visually appealing user interfaces.

OpenCV: OpenCV is employed for image processing. It is used to display uploaded images and highlight the recognized text.

EasyOCR: EasyOCR is a deep learning-based library for OCR. In EyeLight, it plays a pivotal role in accurately recognizing and extracting text from images.

gTTS: gTTS is used for converting extracted text into audio. It facilitates a user-friendly and inclusive experience for those who prefer listening to the content.

Challenges we ran into

During the development of EyeLight, we encountered several challenges that tested our problem-solving and technical skills. One challenge was the OCR process, especially when handling images with complex layouts or low-quality scans. We also faced issues related to user interface design, ensuring that the application remained user-friendly and intuitive. Additionally, the integration of text-to-speech functionality presented their own set of complexities.

Accomplishments that we're proud of

To make an accessible and user-friendly OCR-powered web application that bridges the gap between visual and auditory information

What we learned

I enhanced technical proficiency in Python programming, web development (using Streamlit), computer vision (using OpenCV), deep learning (EasyOCR), and audio processing (gTTS).

What's next for EyeLight

To not just include text to audio but use deep neural network and convert every object detected to audio

Built With

• easyocr
• gtts
• python
• torch