Skip to content

KidStory: Storybook for Kids

KidStory Logo
AI-Powered Interactive Storybooks for Children (Ages 3-10)
Speak a story. Watch it come to life.

Live Demo | Documentation | Architecture

Gemini Live Agent Challenge | Category: Creative Storyteller

"Stop typing, and start interacting." — KidStory moves beyond static chatbots to create an immersive, multimodal storytelling experience powered by Google Gemini.


The Problem: Children Are Losing the Habit of Reading

Children worldwide are spending more time on social media than ever before — and it's replacing reading, creative play, and healthy development.

The Data

Global: U.S. teens spend an average of 4.8 hours per day on social media, with 1.9 hours on YouTube and 1.5 hours on TikTok. Just over half (51%) of teenagers report spending at least four hours daily on social media apps. (Gallup, 2023)

Children specifically: Kids aged 4-18 spend an average of 112 minutes daily on TikTok60% more time than on YouTube. (Qustodio via TechCrunch, 2024)

Southeast Asia (ASEAN): School-aged children (ages 6-14) spend 2.77 hours per day on screens, exceeding the recommended 2-hour limit. In Malaysia, over 90% of adolescents aged 13-17 maintain social media accounts. (HealthcareAsia, 2024)

Very young children: In the UK, an estimated 814,000 children aged 3-5 are actively using social media (37% of that age group), up from 29% in 2023. (Centre for Social Justice, 2024)

Timor-Leste (developer's home country): Over 72% of the population is under 35, with 578,500 social media users (42.2% penetration) and 1.67 million mobile connections (122% of population). Youth actively use Facebook, YouTube, and growing TikTok daily. However, children's book access and reading resources remain extremely limited. (DataReportal, 2024; DiploFoundation)

The Impact on Children

Excessive screen time is not harmless:

  • Developmental delays: A study of 6,000+ children (ages 2-8) found that over 90 minutes of daily screen time is linked to below-average performance in communication, writing, and numeracy (Centre for Social Justice, 2024)
  • Online safety risks: Between 2019-2022, millions of online child abuse cases were reported across Southeast Asia — nearly 8 million in the Philippines, close to 3 million in Vietnam, and 2 million in Thailand (HealthcareAsia, 2024)
  • Approximately one in three internet users globally is a child — yet most digital content children consume is passive scrolling, not educational or creative

The Gap

Children are drawn to screens because the content is engaging — colorful, animated, interactive. But most of that content is passive consumption (watching videos, scrolling feeds). There is a lack of interactive, creative, educational digital experiences that match the engagement level of social media while actually benefiting children's development.


The Solution: KidStory

KidStory transforms screen time from passive consumption into active creation. Instead of watching someone else's content, children speak their own story ideas and watch AI bring them to life with illustrations, narration, and interactive quizzes.

How It Works

Child speaks: "A brave dragon who loves ice cream"
    |
    v
KidStory generates a complete interactive storybook:
    - Story text (4-6 pages)
    - Watercolor illustrations for every page
    - Expressive AI narration
    - Interactive comprehension quiz
    |
    v
Child reads, listens, and learns

Why This Matters

Passive Screen Time (YouTube/TikTok)Active Screen Time (KidStory)
Child watches others' contentChild creates their own stories
No creative inputVoice/text input drives the narrative
No reading practiceFull story text with narration support
No comprehension checkInteractive quiz after each story
Generic contentPersonalized to child's imagination
Potentially unsafe contentBuilt-in child-safety guardrails

Key Features

  • Voice-First Interface — Designed for children who can't type yet
  • Interleaved AI Generation — Story text and illustrations created together in a single Gemini stream
  • Multiple Narrator Voices — Choose from expressive AI personalities
  • Character Consistency — Upload a photo so your child appears in their own story
  • Magic Quiz — AI-generated comprehension questions with voice answers and sound effects
  • Child-Safe Content — Built-in guardrails for ages 3-10

Technical Highlights

KidStory is built for the Gemini Live Agent Challenge Creative Storyteller category, demonstrating:

  • Interleaved Output: Single Gemini call generates both story text and illustrations (responseModalities: ["TEXT", "IMAGE"])
  • Agentic Orchestration: Three specialized Gemini models (Flash Image, Flash TTS, Flash) coordinated by a backend orchestrator
  • Google Cloud Native: Deployed on Cloud Run with Vertex AI, Firestore, Cloud Storage, and Firebase Auth
RequirementImplementation
Gemini Model3 models: gemini-2.5-flash-image, gemini-2.5-flash-preview-tts, gemini-2.5-flash
Google GenAI SDK@google/genai with vertexai: true
Interleaved OutputNative responseModalities: ["TEXT", "IMAGE"] in story generation
Google Cloud HostingCloud Run + Vertex AI + Firestore + Cloud Storage

Architecture

Click here to view the full diagram in a new tab.

KidStory - System Architecture


Project Apps

This repository contains 3 applications:

AppDescriptionURLREADME
KidStory AppMain AI-powered storybook applicationai.kidstory.appstorybook-for-kids-app/README.md
Landing PageMarketing landing page for KidStorykidstory.appstorybook-for-kids-landing/README.md
Documentation SiteFull project documentation sitedocs.kidstory.appstorybook-for-kids-docs/README.md

Documentation

DocumentDescription
App READMEFull application documentation, setup instructions, and tech stack
How It WorksAgentic workflow, model orchestration, step-by-step story and quiz flows
Architecture & DiagramsSystem architecture, sequence diagrams, state machines, data flows
Database SchemaFirestore data models and Cloud Storage layout
Hackathon RequirementsHow KidStory meets every challenge requirement with technical evidence
Deployment GuideStep-by-step Google Cloud Run deployment

Gemini Live Agent Challenge — All required and bonus submission materials:

Required

RequirementLink
Proof of Google Cloud DeploymentProof of Google Cloud deployment/README.md
Architecture DiagramArchitecture & Diagrams

Bonus Points

BonusLink
Blog Post (Max 0.6)Blog Post (GitHub) · Blog Post (dev.to)
Automating Cloud Deployment (Max 0.2)Automating Cloud Deployment/README.md
Google Developer Group Profile (Max 0.2)developers.google.com/profile/u/ajitonelson

Full documentation available at docs.kidstory.app


Getting Started

bash
cd storybook-for-kids-app
npm install
npm run dev

For full setup instructions (Firebase, Vertex AI, environment variables), see the App README.


References

  1. Gallup (2023). Teens Spend Average of 4.8 Hours on Social Media Per Day. news.gallup.com
  2. Qustodio via TechCrunch (2024). Kids Spent 60% More Time on TikTok Than YouTube Last Year. techcrunch.com
  3. HealthcareAsia (2024). Increasing Screen Time Exposes Southeast Asian Children to Online Dangers. healthcareasia.org
  4. Centre for Social Justice (2024). Under 5s Using Social Media. centreforsocialjustice.org.uk
  5. DataReportal (2024). Digital 2024: Timor-Leste. datareportal.com
  6. DiploFoundation. Timor-Leste Digital Policy Overview. dig.watch
  7. The Asia Foundation (2022). Digital Youth in Timor-Leste. asiafoundation.org

Made with care for children everywhere

Released under the MIT License.