Spoke

Audio-first content for creators who speak better than they write.

Record insight → transcribe → format for social → copy & paste.

The Problem

You have insights. You're better at talking than writing. You don't want to learn writing frameworks or reformat the same idea for Twitter, Threads, and Substack.

The Solution

Spoke takes your raw audio, turns it into platform-ready posts in seconds.

v0.0.1 MVP

What it does:

• Record audio (in-app)
• Transcribe (Whisper API)
• Format 3 ways (tweet, thread, long-form) using insight framework
• Copy buttons for each format

What it doesn't do (yet):

• File upload
• Framework selection
• Goal tracking
• Drafts/history
• Social posting
• Auth

Stack

• Frontend: React + Tailwind CSS
• Backend: Supabase
• APIs: Whisper (transcription) + Claude (formatting)
• Hosting: Vercel

Quick Start

git clone https://github.com/YOUR_USERNAME/spoke.git
cd spoke
npm install
npm start

Visit localhost:3000 → record → copy.

Deploy

Push to main → Vercel auto-deploys.

git add .
git commit -m "feature: xyz"
git push origin main

Next (v0.1+)

• File upload
• Framework selection (insight, value, story, question, contrarian)
• Goal tracking (email signups, sales, engagement)
• Drafts/history
• React Native (mobile)
• TestFlight + App Store