Inspiration
While studying I had to write some lengthy boring stuff. Also it's hard to scan the book page and get the summary for writing. It was also hard to revise or memorize all the stuff in textbook. So I build the app to directly get the summary and completion of some text.
What it does
You can upload a text containing picture of textbook or newspaper. this app will extract the text and using cohere API summarize or complete the given text.
How we built it
In backend I used django. I used opencv and pytesseract to extract text from the picture. These libraries use machine learning to extract those texts. Then the text was sent to cohere API to summarize or complete the text. Cohere uses natural language processing to generate summarization and completion. I used html, css and bootstrap templates to do frontend for this project.
Challenges we ran into
I was new to django and API. I had to learn those. It was hard for me to use databases and handling form data with django. In fact I spent the whole night debugging this project.
Accomplishments that we're proud of
Finally the project is completed. I have learnt django. I also learnt to use python AI packages and API's. Which are pretty cool.
What we learned
Django, API, Prompt engineering
What's next for Bonsai
Sentiment analysis, paraphrasing and other important tools can be implemented in bonsai. We can also implement some parameters so that the user can give some commands to get more accurate and user specified output.
Log in or sign up for Devpost to join the conversation.