Robokalam data challenge (Hackathon weeks)
Its not much pleasure to spend an hour or two everyday reading a book and students also face problem in revising contents before the exam as it needs for them to summarize huge texts.
Intent is to build an AI based application that could take input in any document or text format and convert into the way that helps student to have better learning experience.
What it does
- Gets input in the Text, Doc, PDF and Image formats
- Retrieves the text
- Process and summarizes the text
How we built it
- Input and Output: We used streamlit to deploy the application
- Looked at different summarization techniques and found abstract summarization to be most suited for now
Challenges we ran into
- Analyzing and finding the right summarization algorithm
- Finding the right library to read the text and images from pdf
- Processing and summarizing large files
Accomplishments that we're proud of
- We were able to build a decent application that could process few different types of input and summarize them into a pretty smaller text.
- A time saving revision option for students in the competitive digital world
What we learned
- Different algorithms like BART, T5
- Summarizing small and big texts
What's next for Robokalam challenge - Text Summarization
- To process and summarize really large files
- To include Text to speech and Speech to Text
- Also to see if multiple files can be processed at the same time