I built AudioGretel (audiogretel.com), an AI-powered platform that generates personalized audio stories for children in multiple languages.
THE BACKSTORY: As a parent of multilingual kids, I was frustrated with the lack of quality, varied bedtime stories that could also help with language learning. Most story apps either have limited content or are monolingual.
WHAT IT DOES: • Generates unique audio stories using AI (OpenAI GPT) • Supports Spanish, English, Italian, and French • Adapts content for different age groups and language levels • Creates high-quality audio using advanced TTS • Stories are personalized with custom characters and themes
TECHNICAL DETAILS: • React.js frontend with Node.js backend • OpenAI integration for story generation • Multiple TTS engines for natural voices • MongoDB for user data and story history • Optimized for mobile devices • Built with internationalization from day one
INTERESTING CHALLENGES: 1. Balancing AI creativity with child-appropriate content 2. TTS quality varies significantly across languages 3. Optimizing for AI assistant discovery (Gemini, ChatGPT) 4. Managing OpenAI costs while keeping it affordable for families
CURRENT TRACTION: • 10K+ users in 6 months • Featured on Product Hunt • Growing 15% MoM • Freemium model: 3 free stories/month, $5 premium
The response from multilingual families has been amazing. Kids are genuinely learning language patterns through repeated listening, and parents love having fresh content every night.
I'd love feedback from the HN community, especially on: • Technical architecture decisions • AI content generation strategies • Scaling challenges with API-dependent businesses • Ideas for educational integrations
Happy to answer any technical questions about building AI-powered content platforms.
evavillaro•7mo ago