DEV Community

Cover image for How I Built a Storytelling App That Turns Drawings into Tales with Gemini 2.5 Flash
Pavan C
Pavan C

Posted on

How I Built a Storytelling App That Turns Drawings into Tales with Gemini 2.5 Flash

This is a submission for the Google AI Studio Multimodal Challenge

What I Built

I created DoodleTales - a magical storytelling app that transforms children's drawings into interactive stories. Kids can upload any drawing, photo, or artwork, and my app uses Gemini 2.5 Flash to analyze the image and generate personalized 4-page stories with cliffhangers.

The problem I solved is making storytelling more engaging and personalized for children. Instead of generic stories, kids get tales featuring the exact characters and objects from their own drawings. It encourages creativity and makes reading more exciting with interactive "What happens next?" prompts where kids can add their own ideas to continue the story.

Demo

Upload interface with drawing

Upload interface with drawing-1

Upload interface with drawing-2

Generated story pages

Generated story pages

"What happens next?" continuation feature

continuation feature

Story management sidebar

Story management sidebar

How I Used Google AI Studio

I leveraged Google AI Studio to integrate Gemini 2.5 Flash's multimodal capabilities into my Next.js application. The AI Studio made it incredibly easy to:

Test different prompts for image analysis and story generation

Fine-tune the character extraction from drawings

Optimize story structure for kid-friendly content

Debug API responses and improve error handling

The studio's interface helped me experiment with prompt engineering to get the perfect balance of creativity and age-appropriate content.

Multimodal Features

My app uses Gemini 2.5 Flash's image understanding as its core feature:

Visual Analysis: The AI examines uploaded drawings and identifies characters, objects, and themes

Character Extraction: Converts visual elements into story protagonists with child-friendly descriptions

Story Generation: Creates personalized narratives based on what it "sees" in the image

Interactive Continuation: Generates new story content when kids add their own ideas

This multimodal approach transforms static drawings into dynamic, personalized storytelling experiences. Kids feel more connected to stories featuring their own artwork, making reading more engaging and encouraging artistic expression.

Top comments (0)