SpeedRead

Wasting time caught in little details of a paper? Only have two hours for a 20-page paper yet want good annotations? Use SpeedRead! Speed up reading with a new method and improved annotation space.

Comment

Inspiration

We wanted to create something that felt pertinent to our lives as a student. When prompted about efficiency boosters, we realized the inefficacy and need for more convenience of current speed reading tools. Additionally, annotating your readings is crucial, yet many PDFs have tight lines and slim margins. How could we innovate? How could we do better?

What it does

SpeedRead will extract the text from a PDF: It will apply the BionicReading algorithm, a new 'idea' for speed reading. It allows you to customize line spacing, margin width, and font size. Such customization helps annotate by adding more space in the margins and between the lines. SpeedRead includes a simple coaching feature that only allows you to annotate once you finish speed reading. After unlocking a page, you can use the Apple Pencil to annotate. We used the ChatGPT-3.5-Turbo API to suggest lines that are likely worth annotating.

How we built it

We used Swift, UI Kit, Pencil Kit, and PDF Kit.

Challenges we ran into

One of our members is entirely new to Swift and had to learn it on the way on the train. They are also new to the Apple ecosystem. During the first night, they had to download XCode and iOS simulator files, which were over 15 gigabytes. With Harvard's industry-leading and cutting-edge internet speed, this would only take 4 hours. However, the internet would go down, and the downloads would restart from 0%. However, this was a little setback compared to the challenges we faced in implementing the BionicReading algorithm and ensuring that PDFs in an iPad's file system could be imported and exported.

Accomplishments that we're proud of

The team is incredibly proud of what we achieved despite our technical difficulties. The project will be great for our portfolios and resumes! Additionally, the resilience of not just the noobie teammate but all in pushing through frustrating moments as we explored new technologies we were using is something we hope to take away from this experience. And the patience we had to practice during this time of exhaustion and stress is another big takeaway.

What we learned

The main lesson we learned is the importance of planning. Better planning beforehand would have avoided many frustrating moments of annoying bugs and having to refactor or completely rewrite code. Nonetheless, we pivoted well during this project. Another lesson is the importance of incrementing the idea. Instead of starting with something grand, gradually create and improve something that can stand by itself. Also, it was a great moment to develop the chemistry between the members, something we hope to take beyond Harvard.

What's next for SpeedRead

There are many things up next for SpeedRead:

  1. Good zooming feature
  2. Landscape mode
  3. Exporting the PDF with standard page sizes
  4. Adding a column reading mode
  5. ChatGPT being able to explain why parts are important
  6. Select text and link it to an annotation
  7. File and note management/organizing in the app
  8. Be able to tell the difference between headers, body, and footers in the PDF
  9. Improved design

Built With

Share this project:

Updates