TextXtract - AI Text Summarizer







Project Goal 🎯:

Provide a better way for people to get concise, but informative summaries of written content, this had to also include content of different types, whether it's research papers, a long article, legal documents, or even stories. It also had to be able to summarise text written in languages other than English, which would save the user so much time from translating it and then reading it.

Target Audience 📚:

My target audience was clear, is whoever needs to read long pieces of text and understand them, so this meant I had several categories to cover:

1. Students and Academics: For summarizing research papers, articles, and academic texts to quickly grasp key concepts and findings.
2. Professionals and Executives: To condense lengthy reports, business documents, and industry publications for efficient information consumption.
3. Content Creators and Writers: For summarizing existing content to generate outlines or gather insights for new articles and stories.
4. Researchers: In various fields who need to quickly synthesize information from multiple sources.
5. Legal Professionals: For summarizing legal documents, case studies, and precedents to save time and effort.
6. Librarians and Information Specialists: To provide summarized versions of texts for patrons seeking quick information.

Picking the tools for the job 🛠:

Seeing as my application needs were relatively simple, I decided that I needed 2 main things.

• A backend for my text summarization model, for this I chose Flask as it's a lightweight framework, that's easy to set up.
• A frontend to introduce the application to new users, and for them to also be able to generate their summaries. For this, I chose Framer, since Framer offers some awesome design capabilities and also takes way less time than coding it myself. Using code components in Framer, I can easily make requests to my backend to generate summaries.