🎧 MindCast – AI Mood-to-Podcast Generator

💡 What inspired you?

Mental health struggles are at an all-time high, yet most wellness apps require users to manually search through generic content when they're already overwhelmed. I realized people need immediate, personalized support that meets them exactly where they are emotionally – without any barriers or complex interfaces.

🚀 What it does

MindCast transforms your emotional state into personalized wellness podcasts using AI. Simply speak or type how you're feeling, and within seconds, you receive a custom 2-3 minute podcast with breathing exercises, affirmations, and practical guidance tailored to your specific mood.

🛠️ How we built it

• Frontend: Next.js with React for a responsive, accessible interface
• AI Integration: OpenAI GPT-4o via AI SDK for mood analysis and content generation
• Voice Processing: Web Speech API for voice input and browser Text-to-Speech for audio output
• Mood Detection: Custom sentiment analysis that identifies emotions like stressed, anxious, sad, hopeful, etc.
• Data Persistence: Local storage for mood tracking and wellness journey visualization

🎯 Key Features

✅ Voice & Text Input – No typing required when you're struggling✅ Real-time Mood Analysis – AI detects emotional state from natural language✅ Personalized Content – Custom wellness podcasts for each mood✅ Instant Audio Playback – Immediate text-to-speech with calming voices✅ Mood Tracking – Visual history of emotional wellness over time✅ Mobile Responsive – Works seamlessly on any device✅ Offline Capable – Fallback system works without internet

🏔️ Challenges we ran into

• Browser Compatibility: Web Speech API works differently across browsers – implemented robust fallbacks
• AI Response Quality: Ensuring empathetic, therapeutic responses required careful prompt engineering
• Real-time Processing: Balancing speed vs. quality for instant mood analysis
• Audio Synthesis: Making text-to-speech sound natural and calming rather than robotic
• Accessibility: Ensuring the app works for users with different abilities and devices

🎓 What we learned

• AI SDK Integration: How to effectively combine multiple AI models for complex workflows
• Voice Technology: The nuances of browser-based speech recognition and synthesis
• UX for Mental Health: Designing interfaces that reduce friction during emotional distress
• Prompt Engineering: Crafting AI prompts that generate consistently empathetic responses
• Progressive Enhancement: Building apps that gracefully degrade when features aren't available

🏆 What's next for MindCast

• Voice Cloning: Generate podcasts in the user's own voice for deeper personalization
• Ambient Audio Mixing: Blend podcasts with nature sounds, binaural beats, or lo-fi music
• Advanced Analytics: Mood pattern recognition and personalized insights dashboard
• Integration APIs: Connect with Spotify, Apple Health, and meditation apps
• Community Features: Anonymous mood sharing and peer support networks
• Professional Tools: Dashboard for therapists to track client emotional patterns

🌟 Impact & Vision

MindCast democratizes mental health support by providing immediate, personalized care exactly when needed. Our vision is a world where emotional wellness support is as accessible as asking "How are you feeling?" – no appointments, no waiting lists, just instant, empathetic AI companionship available 24/7.

Built With

• ai-sdk
• local
• lucide-react
• next.js-14
• openai-gpt-4o
• radix-ui
• react-18
• shadcn/ui
• speech-synthesis-api
• tailwind-css
• typescript
• vercel
• web-speech-api