Curiofy

Curiofy turns your curiosity into 5-min audio lessons using Gemini + Google TTS. Just type a topic, pick your time, and hit play, AI does the rest. Learn fast, smart, and hands-free.

Comment

Inspiration

In an age of information overload and shrinking attention spans, we wondered: What if learning could be as effortless and addictive as listening to music? That thought sparked Curiofy — an AI-powered microlearning platform that delivers 5-minute podcast-style audio snippets tailored to a user’s interests and time.

We were inspired by the idea of turning idle moments — like waiting in line or commuting — into opportunities for curiosity-driven learning.
The goal: make learning frictionless, personalized, and fun.

What We Built

Curiofy allows users to:

  • Input a topic they’re curious about
  • Set a preferred time limit (e.g., 3, 5, or 7 minutes)
  • Receive an AI-generated, audio-based learning snippet in a natural, conversational tone

Key features include:

  • AI-powered content generation using Gemini
  • Realistic, engaging voiceovers via Google Cloud Text-to-Speech
  • A minimal, intuitive UI for quick, one-step interaction
  • Smart playlist generation for returning users

How We Built It

  • Frontend: React with Tailwind CSS for a fast, responsive user interface
  • Backend: Node.js with Express for API endpoints
  • AI Stack:
    • Gemini API for content generation
    • Google Cloud Text-to-Speech for converting text into human-like audio
  • Database: Supabase for user metadata and session tracking
  • Deployment: Vercel (frontend) and Google Cloud (APIs & TTS)

What We Learned

  • How to fine-tune LLM prompts for concise yet meaningful educational outputs
  • How to integrate cloud-based TTS for smooth and dynamic voice rendering
  • The importance of building a single-click UX that minimizes decision fatigue
  • How curiosity and personalization can boost microlearning engagement

Challenges We Faced

  • Tuning Gemini prompts to balance accuracy with simplicity in content
  • Minimizing audio generation latency for a seamless user experience
  • Designing a voice that sounds natural and avoids the typical TTS monotone
  • Building a content flow that works across any topic, from philosophy to quantum physics

Curiofy continues to evolve with one core belief:

Learning shouldn't feel like a task — it should feel like hitting play on curiosity.

Built With

Share this project:

Updates