Inspiration
- Reading e-books can sometimes lack the vivid imagination and zest that fascinating stories like Harry Potter or Lord of the Rings deserve.
- The Storyteller app aims to enhance the reading experience using advanced technology.
What it does
- GenAI - Text:
- Summarizes the story using an LLM model.
- Lists key learnings from the book.
Takes notes on pages and provides customizable summaries for readers.
GenAI - Images:
Presents caricatures of main characters as described in the book.
Generates scenes of various places mentioned in the book, bringing the story to life.
GenAI - Video:
Creates video summaries with anime or caricatures, providing a unique visual representation.
Text to Speech:
Enables readers to listen to the book using speech-to-text capabilities.
How we built it
- Utilizes multiple base models of GenAI for specific tasks.
- Leverages Langchain or Llama Index wrapper for model integration.
- Implements detailed prompt engineering to achieve desired outputs.
- Integrates seamlessly with Amazon Kindle or any e-book platform.
- Allows users to add books in various formats (e-pub, pdf, documents) for a versatile reading experience.
- Features a user-friendly UI for a seamless experience.
Challenges we ran into
- Availability of books in digital formats.
- Addressing user preferences for physical books.
- Managing token input and output due to multiple features.
Accomplishments that we're proud of
- Unique idea.
- Harnesses GenAI for knowledge spreading.
- Appeals to a wider range of end-users compared to other business cases.
What we learned
- Effective use of GenAI.
- Functionalities of different base models in GenAI.
- Encourages creative thinking.
What's next for Katha-kaar
- Development of mobile versions for wider accessibility.
- Integration of apt base models for enhanced performance.
- Incorporation of GenAI to enable users to discuss, brainstorm, and create their own characters for integration into stories.
- Gathering user feedback for continuous improvement.
Built With
- google-gemini
- google-web-speech-api
- langchain
- neo4j
- openai
- python
- stable-diffussion
- storytelling
- streamlit

Log in or sign up for Devpost to join the conversation.