Inspiration
I got the inspiration from multiple choices based games like quarry, Detroit become human but even they have a fixed storyline and are not readily playable unlike TalesTwist
What it does
It takes in a user prompt (genre, characters, environment...) and generates a story segment with choices to follow along, based on your choice it then proceed the story in that direction
How we built it
I built it using Groq AI and llama LLM model and NextJS and developing appropriate APIs to ensure a smooth user experience. I used Tailwind CSS to make it mobile responsive, light weight and within the hackathon timeframe
Challenges we ran into
- AI Coherence: Crafted specific prompts (e.g., “Generate a 150-word segment with 2 choices”) and post-processed outputs to ensure narrative quality.
- Time Constraint: Built a functional MVP in 48 hours by focusing on core features (prompt input, branching, conclusion).
Accomplishments that we're proud of
A fully responsive high quality UI along with seamless user experience using API and AI Integration Generating a good storyline along with tough choice to encounter
What we learned
Prompt Engineering, AI Integration, API development, Completing a version within a short period of time
What's next for TalesTwist
Integrate Craiyon API for AI-generated scene images to enhance immersion. Add voice input/output using Web Speech API for accessibility. Support multiplayer storytelling for collaborative narratives. Use advanced AI models (GPT-4o, Grok, Claude) Make it multi-modal and even AR-powered to give users a hyper realistic expirence
Built With
- ai
- llm
- nextjs
- node.js
- react
- tailwindcss
- typescript
Log in or sign up for Devpost to join the conversation.