FocusLite

• 

  Logo

Inspiration

The inspiration for FocusLite came from the need to help individuals with ADHD and autism better process and understand complex text. Many existing tools don’t provide enough customization or flexibility for those who need it most. I wanted to create a solution that simplifies reading without overwhelming the user, improving their ability to focus and absorb information.

What it does

FocusLite is a web-based tool that simplifies complex text to improve accessibility and readability for people with ADHD and autism. It provides customizable focus profiles (light, medium, aggressive) that adjust the level of simplification. Additionally, FocusLite extracts relevant content from web pages like news articles and Wikipedia, filtering out distractions such as ads, navigation, and sidebars. The tool also features a text-to-speech function for users who benefit from auditory learning.

How we built it

I built FocusLite using a combination of FastAPI for the backend and HTML, CSS, and JavaScript for the frontend. For the AI text simplification, we leveraged Hugging Face and OpenRouter APIs. The content extraction was implemented using BeautifulSoup to remove unnecessary elements from web pages. The frontend communicates with the backend via API calls to send and retrieve simplified text. I also integrated text-to-speech functionality for additional accessibility.

Challenges we ran into

One of the major challenges was handling content extraction from dynamic, JavaScript-heavy websites. Many sites, like news outlets, have paywalls, which made testing extraction more difficult. Additionally, I faced challenges in fine-tuning the AI text simplification to provide a balance between clarity and accuracy, ensuring it would work effectively across a variety of content types.

Accomplishments that we're proud of

I successfully created a functional tool that simplifies text with multiple focus profiles, ensuring it meets the needs of people with ADHD and autism. The content extraction works reliably across various static websites, and the text-to-speech feature adds an extra layer of accessibility. Additionally, I built a user-friendly interface that’s easy to navigate and can be accessed from any device with a modern web browser.

What we learned

Throughout this project, I learned a lot about AI integration, especially working with third-party APIs for text simplification. We also gained experience in building accessible web applications and how important it is to focus on simplicity in design and functionality. I learned how to handle edge cases in content extraction and optimized for different types of content, such as news articles and Wikipedia pages.

What's next for FocusLite

Next, I plan to expand FocusLite's capabilities by integrating more content sources and improving the AI’s ability to simplify more complex, longer-form text. We also aim to enhance the text-to-speech feature with more natural-sounding voices and provide more customization for the users. Furthermore, I'd like to work on a mobile app version to make FocusLite more accessible across platforms.

Built With

• css
• html
• javascript
• openrouter
• python