EchoShare

EchoShare: When Words Fail, Your Voice Speaks

Comment

EchoShare – Project Story

🚀 Inspiration

In many parts of the world, people still struggle with reading and writing. Whether due to illiteracy, language barriers, or disabilities, millions of individuals are excluded from platforms that rely solely on text. EchoShare was inspired by the belief that everyone has a voice worth hearing—and that voice is sometimes more authentic and impactful than written words. In communities where trust is low, audio adds emotion and credibility that text often lacks.

We wanted to build a platform where spoken words are the primary form of expression, allowing anyone—regardless of literacy level—to share thoughts, opinions, or stories with ease.

🎙 What it does

EchoShare is a social audio platform where users can:

  • Create posts by recording voice or typing (which auto-generates audio via ElevenLabs)
  • Listen to posts shared by others
  • Follow other users and view their profiles
  • Like and comment on posts (via text or voice)
  • Search for users
  • View follower/following lists
  • Experience a clean, responsive UI with accessibility in mind

🔧 How we built it

  • Bolt.new: Used to scaffold the full-stack app with AI assistance for UI, Supabase, and Edge Functions
  • Supabase:
    • Auth for user sign-up/login
    • Post, comments, and followers tables
    • Row-Level Security (RLS) to protect data per user
    • Storage buckets to host audio files
  • ElevenLabs API:
    • Used for converting text posts and text comments into realistic voice
    • Integrated via Supabase Edge Functions to generate audio asynchronously
  • Frontend:
    • React (via Bolt)
    • Responsive feed, profile, and comment components
    • MediaRecorder API for voice input
  • Deployment:
    • Hosted on Netlify
    • Custom domain configured with IONOS through Entri
    • “Powered by Bolt.new” badge placed and responsive

🧱 Challenges we ran into

  • Stuck Audio Generation: Posts showed “Generating audio…” indefinitely due to bulk creation overload. We solved this by:
    • Creating posts asynchronously one at a time
    • Adding audio status fields (pending, done, failed)
    • Deleting broken or hung posts
  • Supabase Storage RLS: Storage was rejecting uploads (403 Unauthorized) due to missing row-level security. We manually wrote RLS policies that:
    • Allow authenticated users to upload their own files
    • Prevent unauthorized access or deletion
  • UUID vs Text Errors: Supabase was throwing type errors on matching UUIDs with text. We fixed it by adding explicit type casts in our SQL queries.
  • Badge Visibility: Bolt.new’s badge was getting hidden under the navigation bar. We updated placement using z-50 and fixed top-4 right-4 with responsive sizing.

🏆 Accomplishments that we're proud of

  • Fully voice-driven social posting workflow (record, upload, or type → generate audio)
  • Voice-based commenting system
  • Clean and accessible design, even for non-readers
  • Follow/follower functionality with search and profile routing
  • Successfully integrated Supabase, ElevenLabs, and Bolt.new into a unified user experience
  • Deployed with custom domain and Netlify

📚 What we learned

  • How to use Supabase RLS for secure user data isolation
  • Best practices for async Edge Function workflows and audio pipelines
  • Audio UX principles, especially around accessibility, fallbacks, and permissions
  • Clean UI/UX design with modern frameworks using AI-assisted tools like Bolt.new
  • The importance of gradual async data seeding vs bulk operations in serverless apps

🔮 What's next for EchoShare

  • Voice-based challenges: Daily themes like “Your Story in 30 Seconds” or “Voice of Hope”
  • Auto-translate audio: Use ElevenLabs multilingual voices to translate posts
  • NFT minting of audio posts: Enable creators to tokenize their work using Algorand
  • Voice cloning for creators: Let verified users clone their voice for custom narration
  • Comment Threads and Mentions
  • Offline recording support for low-bandwidth areas
  • More accessibility options like on-demand transcripts and captions

EchoShare is more than just an app—it’s a platform that gives everyone a voice, literally. Whether you can write or not, your voice matters. Let it be heard. 🌍🎤

Built With

  • algorand-sdk-(optional)
  • bolt.new
  • database
  • edge-functions)
  • elevenlabs-api
  • figma
  • github
  • ionos-(via-entri)
  • mediarecorder-api
  • netlify
  • react
  • reddit
  • stackblitz
  • storage
  • supabase-(auth
  • supabase-integration
  • tailwind-css
  • typescript
Share this project:

Updates