Skip to content

Odzen/sixth

Repository files navigation

Sixth 🎧

Voice AI that helps people with disabilities experience visual content through sound.

Built for the ElevenLabs Voice AI Global Hackathon - Making visual content accessible for the 2.2 billion people worldwide with visual impairments. Took the 1st place in Mexico City.

💡 The Problem

Over 2.2 billion people worldwide have visual impairments. They miss out on visual content every day - from understanding what a sunset looks like to experiencing the colors of a rainbow. Traditional text descriptions and screen readers can't convey the depth, beauty, and emotion of visual experiences.

Sixth solves this with AI-powered spatial audio that transforms visual concepts into immersive sound experiences.

✨ What It Does

  • Accessibility First: Designed specifically for people with visual impairments
  • 🎙️ Voice AI Agent: Real-time conversational agent powered by ElevenLabs
  • 🎵 Spatial Audio: Creates immersive 3D soundscapes that convey depth and emotion
  • 🌅 Visual to Audio: Transforms visual concepts (sunsets, colors, landscapes) into rich audio narratives
  • 🎨 Interactive Demos: Try experiencing "How does the sun feel?" or "What does a rainbow look like?"
  • 🌍 Universal Design: Making content accessible for everyone

🎬 Quick Demo

  1. Visit the live site: sixth-ten.vercel.app
  2. Click "Generate Audio Experience"
  3. Grant microphone permission
  4. Listen as the AI creates an immersive audio description
  5. Speak to continue the conversation!

🚀 Setup

# Install dependencies
npm install

# Add your ElevenLabs Agent ID
echo "NEXT_PUBLIC_ELEVENLABS_AGENT_ID=your-agent-id" > .env.local

# Run locally
npm run dev

Get your Agent ID from elevenlabs.io/app/agents

🛠️ Tech Stack

  • Next.js 15 - React framework (latest)
  • React 19 - UI library (latest)
  • ElevenLabs Agents - Voice AI platform
  • WebRTC - Real-time audio streaming
  • Tailwind CSS 3.4 - Styling
  • Framer Motion - Animations
  • TypeScript 5.7 - Type safety

🎯 Features for Voice AI Hackathon

Assistive Technology - Helps people with disabilities access visual content
Real-time voice interaction - Natural conversational AI agent
WebRTC audio streaming - Low-latency spatial audio for immersive experiences
Accessibility-first design - Built by and for the disability community
Social Impact - Addressing a real problem for 2.2 billion people
Scalable Solution - Works for any visual concept or content

🌐 Live Demo

URL: sixth-ten.vercel.app

Try These Prompts:

  • "How does it feel to see the sun?"
  • "What does a rainbow look like?"
  • "Describe a mountain horizon"

📝 License

© 2025 Sixth - Built for Voice AI Hackathon

About

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published