metadata

title: TECH TALES
emoji: 🚀
colorFrom: red
colorTo: indigo
sdk: gradio
sdk_version: 5.13.2
app_file: app.py
pinned: false
license: apache-2.0
short_description: Bulldog Children's Story Generation

Tech Tales: AI-Powered Story Creation 📚

A creative storytelling application that combines landscape generation, story writing, and image generation to create illustrated children's stories featuring Champ, an adventurous English bulldog.

🌟 Features

1. Landscape Generation

Automatically generates beautiful landscape scenes using Stable Diffusion
Creates unique backgrounds for story inspiration

2. Scene Analysis

Analyzes generated landscapes using SmolVLM
Provides detailed descriptions of the scene elements

3. Story Generation

Creates engaging children's stories using SmolLM2
Features Champ, a lovable English bulldog as the main character
Incorporates themes of confidence, teamwork, caring, and hope
Ensures age-appropriate content with moral lessons

4. Scene Visualization

Generates custom illustrations using a fine-tuned Stable Diffusion model with bulldog LoRA
Creates watercolor-style images that match the story's narrative
Combines text and images for a complete storybook experience

5. Audio Narration

Converts the story into spoken narration
Uses high-quality text-to-speech for natural-sounding storytelling

🎯 How to Use

Click "Generate Random Landscape" to create a starting scene
Get a brief description of the generated landscape
Generate a children's story based on the scene
Create scene prompts from the story
Generate illustrations for each scene
Add text overlays to create storybook pages
Listen to the story being read aloud

🛠️ Technical Details

This space uses several AI models:

Stable Diffusion v1.5 for landscape generation
Custom LoRA adapter for bulldog illustrations
SmolVLM-500M for image analysis
SmolLM2-1.7B for story generation
Kokoro TTS for audio narration

🎨 Custom Theme

The interface features a custom color scheme with:

Light blue primary elements
Light red accent colors
Clean, intuitive layout for easy navigation

📝 Notes

Story generation takes around 30-60 seconds
Each scene illustration takes approximately 30 seconds to generate
The complete process typically takes 3-5 minutes

🙏 Acknowledgments

Stable Diffusion by Runway & CompVis
LoRA bulldog adaptation by Prof-Hunt
Hugging Face for model hosting and Spaces platform

Created with ❤️ using Gradio and Hugging Face Spaces